Stateflow®

Getting Started Guide

R2023b
How to Contact MathWorks

Latest news: www.mathworks.com

Sales and services: www.mathworks.com/sales_and_services

User community: www.mathworks.com/matlabcentral

Technical support: www.mathworks.com/support/contact_us

Phone: 508-647-7000

The MathWorks, Inc.

1 Apple Hill Drive
Natick, MA 01760-2098
Stateflow® Getting Started Guide
© COPYRIGHT 2004–2023 by The MathWorks, Inc.
The software described in this document is furnished under a license agreement. The software may be used or copied
only under the terms of the license agreement. No part of this manual may be photocopied or reproduced in any form
without prior written consent from The MathWorks, Inc.
FEDERAL ACQUISITION: This provision applies to all acquisitions of the Program and Documentation by, for, or through
the federal government of the United States. By accepting delivery of the Program or Documentation, the government
hereby agrees that this software or documentation qualifies as commercial computer software or commercial computer
software documentation as such terms are used or defined in FAR 12.212, DFARS Part 227.72, and DFARS 252.227-7014.
Accordingly, the terms and conditions of this Agreement and only those rights specified in this Agreement, shall pertain
to and govern the use, modification, reproduction, release, performance, display, and disclosure of the Program and
Documentation by the federal government (or other entity acquiring for or through the federal government) and shall
supersede any conflicting contractual terms or conditions. If this License fails to meet the government's needs or is
inconsistent in any respect with federal procurement law, the government agrees to return the Program and
Documentation, unused, to The MathWorks, Inc.
Trademarks
MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See
www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be
trademarks or registered trademarks of their respective holders.
Patents
MathWorks products are protected by one or more U.S. patents. Please see www.mathworks.com/patents for
more information.
Revision History
June 2004 First printing New for Version 6.0 (Release 14)
October 2004 Online only Revised for Version 6.1 (Release 14SP1)
March 2005 Online only Revised for Version 6.2 (Release 14SP2)
September 2005 Online only Revised for Version 6.3 (Release 14SP3)
October 2005 Reprint Version 6.0
March 2006 Second printing Revised for Version 6.4 (Release 2006a)
September 2006 Reprint Version 6.5 (Release 2006b)
March 2007 Online only Rereleased for Version 6.6 (Release 2007a)
September 2007 Third printing Rereleased for Version 7.0 (Release 2007b)
March 2008 Fourth printing Revised for Version 7.1 (Release 2008a)
October 2008 Fifth printing Revised for Version 7.2 (Release 2008b)
March 2009 Sixth printing Revised for Version 7.3 (Release 2009a)
September 2009 Online only Revised for Version 7.4 (Release 2009b)
March 2010 Online only Revised for Version 7.5 (Release 2010a)
September 2010 Online only Revised for Version 7.6 (Release 2010b)
April 2011 Seventh printing Revised for Version 7.7 (Release 2011a)
September 2011 Online only Revised for Version 7.8 (Release 2011b)
March 2012 Online only Revised for Version 7.9 (Release 2012a)
September 2012 Online only Revised for Version 8.0 (Release 2012b)
March 2013 Online only Revised for Version 8.1 (Release 2013a)
September 2013 Online only Revised for Version 8.2 (Release 2013b)
March 2014 Online only Revised for Version 8.3 (Release 2014a)
October 2014 Online only Revised for Version 8.4 (Release 2014b)
March 2015 Online only Revised for Version 8.5 (Release 2015a)
September 2015 Online only Revised for Version 8.6 (Release 2015b)
October 2015 Online only Rereleased for Version 8.5.1 (Release 2015aSP1)
March 2016 Online only Revised for Version 8.7 (Release 2016a)
September 2016 Online only Revised for Version 8.8 (Release 2016b)
March 2017 Online only Revised for Version 8.9 (Release 2017a)
September 2017 Online only Revised for Version 9.0 (Release 2017b)
March 2018 Online only Revised for Version 9.1 (Release 2018a)
September 2018 Online only Revised for Version 9.2 (Release 2018b)
March 2019 Online only Revised for Version 10.0 (Release 2019a)
September 2019 Online only Revised for Version 10.1 (Release 2019b)
March 2020 Online only Revised for Version 10.2 (Release 2020a)
September 2020 Online only Revised for Version 10.3 (Release 2020b)
March 2021 Online only Revised for Version 10.4 (Release 2021a)
September 2021 Online only Revised for Version 10.5 (Release 2021b)
March 2022 Online only Revised for Version 10.6 (Release 2022a)
September 2022 Online only Revised for Version 10.7 (Release 2022b)
March 2023 Online only Revised for Version 10.8 (Release 2023a)
September 2023 Online only Revised for Version 23.2 (R2023b)
 Contents

Introduction to the Stateflow Product

1
Stateflow Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

Stateflow Fundamentals
2
Model Finite State Machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Example of a Stateflow Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Execute Chart as a MATLAB Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
Simulate Chart as a Simulink Block with Local Events . . . . . . . . . . . . . . . 2-4
Simulate Chart as a Simulink Block with Temporal Conditions . . . . . . . . . 2-7
Next Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9

Construct and Run a Stateflow Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10

Construct the Stateflow Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Simulate the Chart as a Simulink Block . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
Execute the Chart as a MATLAB Object . . . . . . . . . . . . . . . . . . . . . . . . . 2-15

Define Chart Behavior by Using State Actions and Transition Labels . . 2-17

Create a Hierarchy to Manage System Complexity . . . . . . . . . . . . . . . . . . 2-21

Model Synchronous Subsystems by Using Parallel Decomposition . . . . 2-25

Synchronize Parallel States by Broadcasting Events . . . . . . . . . . . . . . . . 2-29

Monitor Chart Activity by Using Active State Data . . . . . . . . . . . . . . . . . . 2-36

Schedule Chart Actions by Using Temporal Logic . . . . . . . . . . . . . . . . . . 2-42

Additional Learning Tools

3

v
 1

Introduction to the Stateflow Product

1 Introduction to the Stateflow Product

Stateflow Product Description

Model and simulate decision logic using state machines and flow charts

Stateflow provides a graphical language that includes state transition diagrams, flow charts, state
transition tables, and truth tables. You can use Stateflow to describe how MATLAB® algorithms and
Simulink® models react to input signals, events, and time-based conditions.

Stateflow enables you to design and develop supervisory control, task scheduling, fault management,
communication protocols, user interfaces, and hybrid systems.

With Stateflow, you model combinatorial and sequential decision logic that can be simulated as a
block within a Simulink model or executed as an object in MATLAB. Graphical animation enables you
to analyze and debug your logic while it is executing. Edit-time and run-time checks ensure design
consistency and completeness before implementation.

1-2
 2

Stateflow Fundamentals

This chapter describes the fundamental concepts of event-based modeling in Stateflow.

• “Model Finite State Machines” on page 2-2

• “Construct and Run a Stateflow Chart” on page 2-10
• “Define Chart Behavior by Using State Actions and Transition Labels” on page 2-17
• “Create a Hierarchy to Manage System Complexity” on page 2-21
• “Model Synchronous Subsystems by Using Parallel Decomposition” on page 2-25
• “Synchronize Parallel States by Broadcasting Events” on page 2-29
• “Monitor Chart Activity by Using Active State Data” on page 2-36
• “Schedule Chart Actions by Using Temporal Logic” on page 2-42
2 Stateflow Fundamentals

Model Finite State Machines

Stateflow is a graphical programming environment based on finite state machines. With Stateflow,
you can test and debug your design, consider different simulation scenarios, and generate code from
your state machine.

Finite state machines are representations of dynamic systems that transition from one mode of
operation (state) to another. State machines:

• Serve as a high-level starting point for a complex software design process.

• Enable you to focus on the operating modes and the conditions required to pass from one mode to
the next mode.
• Help you to design models that remain clear and concise even as the level of model complexity
increases.

Control systems design relies heavily on state machines to manage complex logic. Applications
include designing aircraft, automobiles, and robotics control systems.

Example of a Stateflow Chart

In a Stateflow chart, you combine states, transitions, and data to implement a finite state machine.
This Stateflow chart presents a simplified model of the logic to shift gears in a four-speed automatic
transmission system of a car. The chart represents each gear position by a state, shown as a rectangle
labeled first, second, third, or fourth. Like the gears they represent, these states are exclusive,
so only one state is active at a time.

The arrow on the left of the diagram represents the default transition and indicates the first state to
become active. When you execute the chart, this state is highlighted on the canvas. The other arrows
indicate the possible transitions between the states. To define the dynamics of the state machine, you
associate each transition with a Boolean condition or a trigger event. For example, this chart
monitors the speed of the car and shifts to a different gear when the speed crosses a fixed threshold.
During simulation, the highlighting in the chart changes as different states become active.

This chart offers a simple design that disregards important factors such as engine speed and torque.
You can construct a more comprehensive and realistic model by linking this Stateflow chart with
other components in MATLAB or Simulink. The following examples describe three possible
approaches.

2-2
 Model Finite State Machines

Execute Chart as a MATLAB Object

This example presents a modified version of an automatic transmission system that incorporates state
hierarchy, temporal logic, and input events.

• Hierarchy: The chart consists of a superstate gear_logic that surrounds the four-speed
automatic transmission chart in the previous example. This superstate controls the speed and
acceleration of the car. During execution, gear_logic is always active.
• Temporal Logic: In the state gear_logic, the action on every(0.25,sec) determines the
speed of the car. The operator every creates a MATLAB timer that executes the chart and
updates the chart data speed every 0.25 seconds.
• Input Events: The input events SpeedUp, Cruise, and SlowDown reset the value of the chart
data delta. This data determines whether the car accelerates or maintains its speed at each
execution step.

You can execute this chart as an object in MATLAB directly through the Command Window or by
using a script. You can also program a MATLAB app that controls the state of the chart through a
graphical user interface. For example, this user interface sends an input event to the chart when you
click a button. In the chart, the MATLAB function widgets controls the values of the gauges and
lamps on the interface. To start the example, in the App Designer toolstrip, click Run. The example
continues to run until you close the user interface window.

2-3
2 Stateflow Fundamentals

Alternatively, in the Stateflow Editor, in the State Chart Tab, click Run. To control the speed of the
car, in the Symbols pane, use the SpeedUp, SlowDown, and Cruise buttons. To stop the example,
click Stop.

For more information about executing Stateflow charts as MATLAB objects, see “Execution in
MATLAB”.

Simulate Chart as a Simulink Block with Local Events

This example provides a more complex design for an automatic transmission system. The Stateflow
chart appears as a block in a Simulink model. The other blocks in the model represent related
automotive components. The chart interfaces with the other blocks by sharing data through input and
output connections. To open the chart, click the arrow in the bottom left corner of the shift_logic
block.

This chart combines state hierarchy, parallelism, active state data, local events, and temporal logic.

2-4
 Model Finite State Machines

• Hierarchy: The state gear_state contains a modified version of the four-speed automatic
transmission chart. The state selection_state contains substates that represent the steady
state, upshifting, and downshifting modes of operation. When circumstances require a shift to a
higher or lower gear, these states become active.
• Parallelism: The parallel states gear_state and selection_state appear as rectangles with
a dashed border. These states operate simultaneously, even as the substates inside them turn on
and off.
• Active State Data: The output value gear reflects the choice of gears during simulation. The
chart generates this value from the active substate in gear_state.
• Local Events: In place of Boolean conditions, this chart uses the local events UP and DOWN to
trigger the transitions between gears. These events originate from the send commands in
selection_state when the speed of the car goes outside the range of operation for the selected
gear. The Simulink function calc_th determines the boundary values for the range of operation
based on the selected gear and the engine speed.
• Temporal Logic: To prevent a rapid succession of gear changes, selection_state uses the
temporal logic operator after to delay the broadcasting of the UP and DOWN events. The state
broadcasts one of these events only if a change of gears is required for longer than some
predetermined time TWAIT.

To run a simulation of the model:

1 Double-click the User Inputs block. In the Signal Editor dialog box, select a predefined brake-to-
throttle profile from the Active Scenario list. The default profile is Passing Maneuver.

2-5
2 Stateflow Fundamentals

2 Click Run. In the Stateflow Editor, chart animation highlights the active states during the
simulation. To slow down the animation, in the Debug tab, select Slow from the Animation
Speed drop-down list.
3 In the Scope blocks, examine the results of the simulation. Each scope displays a graph of its
input signals during simulation.

2-6
 Model Finite State Machines

Simulate Chart as a Simulink Block with Temporal Conditions

This example provides another alternative for modeling the transmission system in a car. The
Stateflow chart appears as a block in a Simulink model. The other blocks in the model represent
related automotive components. The chart interfaces with the other blocks by sharing data through
input and output connections. To open the chart, click the arrow in the bottom left corner of the
Gear_logic block.

This chart combines state hierarchy, active state data, and temporal logic.

2-7
2 Stateflow Fundamentals

• Hierarchy: This model places the four-speed automatic transmission chart inside a superstate
gear. The superstate monitors the vehicle and engine speeds and triggers gear changes. The
actions listed on the upper left corner of the state gear determine the operating thresholds for the
selected gear and the values of the Boolean conditions up and down. The label en,du indicates
that the state actions are executed when the state first becomes active (en = entry) and at every
subsequent time step while the state is active (du = during).
• Active State Data: The output value gear reflects the choice of gears during simulation. The
chart generates this value from the active substate in gear.
• Temporal Logic: To prevent a rapid succession of gear changes, the Boolean conditions up and
down use the temporal logic operator duration to control the transition between gears. The
conditions are valid when the speed of the car remains outside the range of operation for the
selected gear longer than some predetermined time TWAIT (measured in seconds).

To run a simulation of the model:

1 Double-click the User Inputs block. In the Signal Editor dialog box, select a predefined brake-to-
throttle profile from the Active Scenario list. The default profile is Passing Maneuver.
2 Click Run. In the Stateflow Editor, chart animation highlights the active states during the
simulation. To slow down the animation, in the Debug tab, select Slow from the Animation
Speed drop-down list.
3 In the Scope block, examine the results of the simulation. The scope displays a graph of the gear
selected during simulation.

2-8
 Model Finite State Machines

Next Steps
1 “Construct and Run a Stateflow Chart” on page 2-10
2 “Define Chart Behavior by Using State Actions and Transition Labels” on page 2-17
3 “Create a Hierarchy to Manage System Complexity” on page 2-21
4 “Model Synchronous Subsystems by Using Parallel Decomposition” on page 2-25
5 “Synchronize Parallel States by Broadcasting Events” on page 2-29
6 “Monitor Chart Activity by Using Active State Data” on page 2-36
7 “Schedule Chart Actions by Using Temporal Logic” on page 2-42

2-9
2 Stateflow Fundamentals

Construct and Run a Stateflow Chart

A Stateflow chart is a graphical representation of a finite state machine consisting of states,
transitions, and data. You can create a Stateflow chart to define how a MATLAB algorithm or a
Simulink model reacts to external input signals, events, and time-based conditions.

For instance, this Stateflow chart presents the logic underlying a half-wave rectifier. The chart
contains two states labeled On and Off. In the On state, the chart output signal y is equal to the input
x. In the Off state, the output signal is set to zero. When the input signal crosses some threshold t0,
the chart transitions between these states. The actions in each state update the value of y at each
time step of the simulation.

This example shows how to create this Stateflow chart for simulation in Simulink and execution in
MATLAB.

Construct the Stateflow Chart

Open the Stateflow Editor

The Stateflow Editor is a graphical environment for designing state transition diagrams, flow charts,
state transition tables, and truth tables. Before opening the Stateflow Editor, decide on the chart
execution mode that best meets your needs.

• To model conditional, event-based, and time-based logic for periodic or continuous-time Simulink
algorithms, use the sfnew function to create a Stateflow chart that you can simulate as a block in
a Simulink model. At the MATLAB command prompt, enter:

sfnew rectify % create chart for simulation in a Simulink model

Simulink creates a model called rectify that contains an empty Stateflow Chart block. To open
the Stateflow Editor, double-click the chart block.
• To design reusable state machine and timing logic for MATLAB applications, use the edit
function to create a standalone Stateflow chart that you can execute as a MATLAB object. At the
MATLAB command prompt, enter:

edit rectify.sfx % create chart for execution as a MATLAB object

If the file rectify.sfx does not exist, the Stateflow Editor creates an empty chart with the name
rectify.

The main components of the Stateflow Editor are the chart canvas, the object palette, and the
Symbols pane.

• The chart canvas is a drawing area where you create a chart by combining states, transitions, and
other graphical elements.

2-10
 Construct and Run a Stateflow Chart

• On the left side of the canvas, the object palette displays a set of tools for adding graphical
elements to your chart.
• On the right side of the canvas, in the Symbols pane, you add new data, events, and messages to
the chart and resolve any undefined or unused symbols.

Tip After you construct your Stateflow chart, you can copy its contents to another chart with a
different execution mode. For example, you can construct a chart for execution in MATLAB and copy
its contents into a chart for simulation in Simulink.

Add States and Transitions

1
From the object palette, click the State icon and move the pointer to the chart canvas. A
state with its default transition appears. To place the state, click a location on the canvas. At the
text prompt, enter the state name On and the state action y = x.

2 Add another state. Right-click and drag the On state. Blue graphical cues help you to align your
states horizontally or vertically. The name of the new state changes to Off. Double-click the state
and modify the state action to y = 0.

2-11
2 Stateflow Fundamentals

3 Realign the two states and pause on the space between the two states. Blue transition cues
indicate several ways in which you can connect the states. To add transitions, click the
appropriate cue.

Alternatively, to draw a transition, click and drag from the edge of one state to the edge of the
other state.

4 Double-click each transition and type the appropriate transition condition x<t0 or x>=t0. The
conditions appear inside square brackets.

5 Clean up the chart:

• To improve clarity, move each transition label to a convenient location above or below its
corresponding transition.
• To align and resize the graphical elements of your chart, in the Format tab, click Auto
Arrange or press Ctrl+Shift+A.
•
To resize the chart to fit the canvas, press the space bar or click the Fit To View icon .

Resolve Undefined Symbols

Before you can execute your chart, you must define each symbol that you use in the chart and specify
its scope (for example, input data, output data, or local data). In the Symbols pane, undefined
symbols are marked with a red error badge . The Type column displays the suggested scope for
each undefined symbol based on its usage in the chart.
1 Open the Symbols pane.

• If you are building a chart in a Simulink model, in the Modeling tab, under Design Data,
select Symbols Pane.
• If you are building a standalone chart for execution in MATLAB, in the State Chart tab, select
Add Data > Symbols Pane.
2
In the Symbols pane, click Resolve Undefined Symbols .

• If you are building a chart in a Simulink model, the Stateflow Editor resolves the symbols x
and t0 as input data and y as output data .
• If you are building a standalone chart for execution in MATLAB, the Stateflow Editor resolves
t0, x, and y as local data .

2-12
 Construct and Run a Stateflow Chart

3 Because the threshold t0 does not change during simulation, change its scope to constant data.
In the Type column, click the data type icon next to t0 and select Constant Data.
4 Set the value for the threshold t0. In the Value column, click the blank entry next to t0 and
enter a value of 0.
5 Save your Stateflow chart.

Your chart is now ready for simulation in Simulink or execution in MATLAB.

Simulate the Chart as a Simulink Block

To simulate the chart inside a Simulink model, connect the chart block to other blocks in the model
through input and output ports. If you want to execute the chart from the MATLAB Command
Window, see “Execute the Chart as a MATLAB Object” on page 2-15.

1 To return to the Simulink Editor, on the explorer bar at the top of the canvas, click the name of
the Simulink model: rectify. If the explorer bar is not visible, click the Hide/Show
Explorer Bar icon at the top of the object palette.
2 Add a source to the model:

• From the Simulink Sources library, add a Sine Wave block.

• Double-click the Sine Wave block and set the Sample time to 0.2.
• Connect the output of the Sine Wave block to the input of the Stateflow chart.
• Label the signal as x.
3 Add a sink to the model:

• From the Simulink Sinks library, add a Scope block with two input ports.
• Connect the output of the Sine Wave block to the first input of the Scope block.
• Connect the output of the Stateflow chart to the second input of the Scope block.
• Label the signal as y.
4 Save the Simulink model.

2-13
2 Stateflow Fundamentals

5
To simulate the model, click Run . During the simulation, the Stateflow Editor highlights
active states and transitions through chart animation.
6 After you simulate the model, double-click the Scope block. The scope displays the graphs of the
input and output signals to the charts.

The simulation results show that the rectifier filters out negative input values.

2-14
 Construct and Run a Stateflow Chart

Execute the Chart as a MATLAB Object

To execute the chart in the MATLAB Command Window, create a chart object and call its step
function. If you want to simulate the chart inside a Simulink model, see “Simulate the Chart as a
Simulink Block” on page 2-13.

1 Create a chart object r by using the name of the sfx file that contains the chart definition as a
function. Specify the initial value for the chart data x as a name-value pair.

r = rectify(x=0);
2 Initialize input and output data for chart execution. The vector X contains input values from a
sine wave. The vector Y is an empty accumulator.

T = 0:0.2:10;
X = sin(T);
Y = [];
3 Execute the chart object by calling the step function multiple times. Pass individual values from
the vector X as chart data x. Collect the resulting values of y in the vector Y. During the
execution, the Stateflow Editor highlights active states and transitions through chart animation.

for i = 1:51
step(r,x=X(i));
Y(i) = r.y;
end
4 Delete the chart object r from the MATLAB workspace.

delete(r)
5 Examine the results of the chart execution. For example, you can call the stairs function to
create a stairstep graph that compares the values of X and Y.

ax1 = subplot(2,1,1);
stairs(ax1,T,X,color="#0072BD")
title(ax1,"x")

ax2 = subplot(2,1,2);
stairs(ax2,T,Y,color="#D95319")
title(ax2,"y")

2-15
2 Stateflow Fundamentals

The execution results show that the rectifier filters out negative input values.

See Also

More About
• “Stateflow Editor Operations”
• “Manage Symbols in the Stateflow Editor”
• “Create Stateflow Charts for Execution as MATLAB Objects”

2-16
 Define Chart Behavior by Using State Actions and Transition Labels

Define Chart Behavior by Using State Actions and Transition

Labels

State actions and transition actions are instructions that you write inside a state or on a transition,
respectively, to define how a Stateflow® chart behaves during simulation. For example, the actions in
this chart define a state machine that empirically verifies one instance of the Collatz conjecture. For a
given numeric input , the chart computes the hailstone sequence … by iterating
this rule:

• If is even, then .
• If is odd, then .

The Collatz conjecture states that every positive integer has a hailstone sequence that eventually
reaches one.

The chart consists of three states. At the start of simulation, the Init state initializes the chart data
by setting:

• Local data n to the value of the input u.

• Local data n2 to the remainder when n is divided by two.
• Output data y to false.

Depending on the parity of the input, the chart transitions to either the Even or Odd state. As the
state activity shifts between the Even and Odd states, the chart computes the numbers in the
hailstone sequence. When the sequence reaches a value of one, the output data y becomes true and
triggers a Stop Simulation (Simulink) block in the Simulink® model.

State Action Types

State actions define what a Stateflow chart does while a state is active. The most common types of
state actions are entry, during, and exit actions:

2-17
2 Stateflow Fundamentals

• entry actions occur when the state becomes active.

• during actions occur on a time step when the state is already active and the chart does not
transition out of the state.
• exit actions occur when the chart transitions out of the state.

You can specify the type of a state action by using a complete keyword (entry, during, exit) or an
abbreviation (en, du, ex). You can also combine state action types by using commas. For instance, an
action with the combined type entry, during occurs on the time step when the state becomes
active and on every subsequent time step while the state remains active.

The hailstone chart contains actions in these states:

• Init — When this state becomes active at the start of the simulation, the entry action
determines the parity of n and sets y to false. When the chart transitions out of Init after one
time step, the exit action determines whether n is equal to one.
• Even — When this state becomes active, and on every subsequent time step that the state is
active, the combined entry, during action computes the value and parity for next number of
the hailstone sequence, n/2.
• Odd — When this state becomes active, and on every subsequent time step that the state is active,
the combined entry, during action checks whether n is greater than one and, if it is, computes
the value and parity for next number of the hailstone sequence, 3*n+1.

Transition Label Types

Transition labels define what a Stateflow chart does when the active state changes. The most
common types of transition labels are conditions and condition actions.
[Condition]{ConditionAction}

Condition is a Boolean expression that determines whether the transition occurs. If you do not
specify a condition, the transition occurs one time step after the source state becomes active.

ConditionAction is an instruction that executes when the condition that guards the transition is
true. The condition action takes place after the condition but before any exit or entry state actions.

The hailstone chart contains actions on these transitions:

• Default transition into Init — At the start of the simulation, the condition action n = u assigns
the input value u to the local data n.
• Transition from Init to Even — The condition n2 == 0 determines that the transition occurs
when n is even. The number 1 at the source of this transition indicates that this transition is
evaluated before the transition Init to Odd.
• Transition from Odd to Even — The condition n2 == 0 determines that the transition occurs
when n is even.
• Transition from Even to Odd — The condition n2 ~= 0 determines that the transition occurs
when n is odd. In this case, the condition action y = isequal(n,1) determines whether n is
equal to one.

Examine Chart Behavior

To compute the hailstone sequence starting with a value of nine:

1. In the Constant block, enter a value of 9.

2-18
 Define Chart Behavior by Using State Actions and Transition Labels

2. In the Simulation tab, click Run. The chart responds with these actions:

• At time , the default transition to Init occurs. The transition action sets the value of n to 9.
The state Init becomes active. The entry actions in Init set n2 to 1 and y to false.
• At time , the condition n2 == 0 is false so the chart prepares to transition to Odd. The exit
action in Init sets y to false. The state Init becomes inactive and the state Odd becomes
active. The entry action in Odd sets n to 28 and n2 to 0.
• At time , the condition n2 == 0 is true so the chart prepares to transition to Even. The state
Odd becomes inactive and the state Even becomes active. The entry action in Even sets n to 14
and n2 to 0.
• At time , the condition n2 ~= 0 is false so the chart does not take a transition. The state
Even remains active. The during action in Even sets n to 7 and n2 to 1.
• At time , the condition n2 ~= 0 is true so the chart prepares to transition to Odd. The
transition action sets y to false. The state Even becomes inactive and the state Odd becomes
active. The entry actions in Odd set n to 22 and n2 to 0.
• The chart continues to compute the hailstone sequence until it arrives at a value of n at time
.
• At time , the chart prepares to transition from Even to Odd. The transition action sets y to
true. The state Even becomes inactive and the state Odd becomes active. The entry actions in
Odd do not modify n or n2. The Stop Simulation block connected to the output signal y stops the
simulation.

3. In the Simulation tab, under Review Results, click Data Inspector.

4. To see the values of the hailstone sequence, in the Simulation Data Inspector, select the logged
signal n.

2-19
2 Stateflow Fundamentals

The hailstone sequence reaches a value of one after 19 iterations.

See Also

More About
• “Represent Operating Modes by Using States”
• “Define Actions in a Transition”

2-20
 Create a Hierarchy to Manage System Complexity

Create a Hierarchy to Manage System Complexity

To control multiple levels of complexity in your system, create a hierarchy of nested states in your
Stateflow® chart by placing one or more states inside the boundaries of another state.

The outer state is the parent, or superstate, of the inner states. The inner states are child states, or
substates, of the outer state. The contents of the superstate behave like a smaller chart. When the
superstate becomes active, one of the substates also becomes active. When the superstate becomes
inactive, all of the substates become inactive.

Model a Media Player

This example models a media system that consists of an FM radio and an internet stream player.
During simulation, you control the media player by interacting with the buttons and knobs on the
Media Player App user interface.

To run the media player:

2-21
2 Stateflow Fundamentals

1 Open the Simulink model and click Run. The Media Player App opens. The media player is
initially off. At the top of the app, the Media Player Status box shows the message Standby
(Off).
2 Turn the Component Selection knob to Stream. The status message briefly displays Connecting
to Handel's Greatest Hits. After a short pause, the status message changes to Playing:
Handel's Greatest Hits and music begins to play.
3 Click the Fast-Forward button. The music stops and chirping sounds begin. The status message
changes to Forward >> Handel's Greatest Hits. The name of the stream scrolls forward
across the display. To resume normal play mode, click the Play button.
4 Click the Reverse button. Chirping sounds play and the status message changes to Reverse >>
Handel's Greatest Hits. The name of the stream scrolls backward across the display. To
resume normal play mode, click the Play button.
5 In the Stream Name box, enter the name of a new stream and click Connect. For example, try
connecting to the streams Training Deep Networks or Fun With State Machines.
6 Turn the Component Selection knob to Radio. The status message displays Playing: 99.5 FM.
To select another station, turn the FM Radio Station knob.
7 To stop the simulation, close the Media Player App.

Implement Behavior by Using State Hierarchy

This example implements the media player by focusing on a single level of activity at a time. For
example, these conditions are necessary for the stream player to enter fast-forward playback mode:

1 You turn on the media player.

2 You select the stream player.
3 You start playing a stream.
4 You click the Fast-Forward button.

The model uses a hierarchy of nested states to consider each condition individually. For instance, the
Model Explorer shows the hierarchy of states in the Mode Manager chart. To open the Model
Explorer, in the Modeling tab, select Model Explorer.

At the top or outermost level of the hierarchy, the Mode Manager chart has two states that control
the activity of the media player:

• Off is active when you turn off the media player.

2-22
 Create a Hierarchy to Manage System Complexity

• On is active when you set the media player to streaming or radio mode.

The child states of On control the media player components:

• Stream is active when you select the stream player.

• Radio is active when you select the FM radio.

The child states of Stream control the activity of the stream player:

• Play is active when you play a stream.

• Pause is active when you pause the stream player.

The child states of Play control the play modes for the stream player:

• Normal is active during normal play mode.

• Reverse is active when you click the Reverse button.
• FastForward is active when you click the Fast-Forward button.

This figure shows the layout of the states in the chart.

Explore the Example

The model in this example contains two other Stateflow charts:

• App Interface manages the interface with the MATLAB app and passes inputs to the Mode
Manager and Stream Player charts.

2-23
2 Stateflow Fundamentals

• Stream Player receives the output from the App Interface and Mode Manager charts and
simulates the internal behavior of the stream player.

During simulation, you can investigate how each chart responds to interactions with the Media Player
App. To switch quickly between charts, use the tabs at the top of the Stateflow Editor.

See Also

More About
• “Use State Hierarchy to Design Multilevel State Complexity”

2-24
 Model Synchronous Subsystems by Using Parallel Decomposition

Model Synchronous Subsystems by Using Parallel

Decomposition

To implement operating modes that run concurrently, use parallel states in your Stateflow® chart.
For example, as part of a complex system design, you can employ parallel states to model
independent components or subsystems that are active simultaneously.

State Decomposition

The decomposition type of a chart or state specifies whether the chart or state contains exclusive
states or parallel states:

• Exclusive states represent mutually exclusive modes of operation. No two exclusive states at the
same hierarchical level can be active or execute at the same time. Stateflow charts represent each
exclusive state by a solid rectangle.

• Parallel states represent independent modes of operation. Two or more parallel states can be
active at the same time, although they execute in a serial fashion. Stateflow charts represent each
parallel state by a dashed rectangle with a number that indicate the execution order.

You can combine exclusive and parallel states in your Stateflow chart by setting the state
decomposition at different levels of your state hierarchy. The default state decomposition type is
Exclusive (OR). To change the decomposition type to Parallel (AND), right-click the parent
state and select Decomposition > Parallel (AND). To change the decomposition type back to
Exclusive (OR), right-click the parent state and select Decomposition > Exclusive (OR).

Model an Air Temperature Controller

This example uses parallel decomposition to model an air controller that maintains air temperature at
120 degrees in a physical plant.

2-25
2 Stateflow Fundamentals

At the top level, the air controller chart has two exclusive states, PowerOff and PowerOn. The chart
uses exclusive (OR) decomposition because controller cannot be on and off at the same time.

The controller operates two fans. The first fan turns on when the air temperature rises above 120
degrees. The second fan provides additional cooling when the air temperature rises above 150
degrees. The chart models these fans as parallel substates, FAN1 and FAN2, of the top-level state
PowerOn. Because the fans operate as independent components that turn on or off depending on how
much cooling is required, PowerOn uses parallel (AND) decomposition to ensure that both substates
are active when the controller is turned on.

Except for the operating thresholds, the fans are modeled by states with an identical configuration of
substates and transitions that reflects the two modes of fan operation, On and Off. Because neither
fan can be on and off at the same time, FAN1 and FAN2 have exclusive (OR) decomposition.

In PowerOn, a third parallel state called SpeedValue represents an independent subsystem that
calculates the number of fans that have cycled on at each time step. The Boolean expression
in(FAN1.On) has a value of 1 when the On state of FAN1 is active. Otherwise, in(FAN1.On) equals
0. Similarly, the value of in(FAN2.On) represents whether FAN2 has cycled on or off. The sum of
these expressions indicates the number of fans that are turned on during each time step.

Specify Order of Execution for Parallel States

Although FAN1, FAN2, and SpeedValue are active concurrently, these states execute in serial fashion
during simulation. The numbers in the upper-right corners of the states specify the order of
execution. The rationale for this order of execution is:

• FAN1 executes first because it cycles on at a lower temperature than FAN2. It can turn on
regardless of whether FAN2 is on or off.

2-26
 Model Synchronous Subsystems by Using Parallel Decomposition

• FAN2 executes second because it cycles on at a higher temperature than FAN1. It can turn on only
if FAN1 is already on.
• SpeedValue executes last so it can observe the most up-to-date status of FAN1 and FAN2.

By default, Stateflow assigns the execution order of parallel states based on the order you add them
to the chart. To change the execution order of a parallel state, right-click the state and select a value
from the Execution Order drop-down list.

Explore the Example

This example contains a Stateflow chart called Air Controller and a Simulink® subsystem called
Physical Plant.

Based on the air temperature of the physical plant, the chart turns on the fans and outputs the
number of running fans, airflow, to the subsystem. This value determines the cooling activity factor,
, according to these rules:

• airflow = 0 — No fans are running. The air temperature does not decrease because .
• airflow = 1 — One fan is running. The air temperature decreases according to the cooling
activity factor .
• airflow = 2 — Two fans are running. The air temperature decreases according to the cooling
activity factor .

The Physical Plant subsystem updates the air temperature, , inside the plant based on the
equations

where:

• is the initial temperature. The default value is 70°.

• is the ambient temperature. The default value is 160°.

2-27
2 Stateflow Fundamentals

• is the heat transfer factor for the plant. The default value is 0.01.
• is the cooling activity factor that corresponds to airflow.

The new temperature determines the amount of cooling at the next time step of the simulation.

See Also

More About
• “Define Exclusive and Parallel Modes by Using State Decomposition”
• “Execution Order for Parallel States”
• “Check State Activity by Using the in Operator”

2-28
 Synchronize Parallel States by Broadcasting Events

Synchronize Parallel States by Broadcasting Events

Local events enable you to coordinate parallel states by allowing one state to trigger a transition or
an action in another state in the same Stateflow® chart. To broadcast an event from one state to
another state, use the send operator with the name of the event and the name of an active state:

send(eventName,stateName)

When you broadcast an event, the event takes effect in the receiving state and in any substates in the
hierarchy of that state.

Model a Home Security System

This example uses local events as part of the design of a home security system.

2-29
2 Stateflow Fundamentals

The security system consists of an alarm and three anti-intrusion sensors: a door sensor, a window
sensor, and a motion detector. After the system detects an intrusion, you have a small amount of time
to disable the alarm. Otherwise, the system calls the police.

The chart models each sensor by using one of these parallel states:

• The parallel state Door models the door sensor. The input signal D_mode selects between the
Active and Disabled modes for this sensor. When the sensor is active, the input signal
Door_sens indicates a possible intrusion.
• The parallel state Win models the window sensor. The input signal W_mode selects between the
Active and Disabled modes for this sensor. When the sensor is active, the input signal
Win_sens indicates a possible intrusion.

2-30
 Synchronize Parallel States by Broadcasting Events

• The parallel state Motion models the motion detector. The input signal M_mode selects between
the Active and Disabled modes for this sensor. When the sensor is active, the input signal
Mot_sens indicates a possible intrusion.

To mitigate the effect of sporadic false positives, the motion detector incorporates a debouncing
design so that only a sustained positive trigger signal produces an alert. In contrast, the door and
window sensors interpret a single positive trigger signal as an intrusion and issue an immediate alert.

A fourth parallel state called Alarm models the operating modes of the alarm system. The input
signal Alarm_active selects between the On and Off modes for the alarm. If a sensor detects an
intrusion while the alarm subsystem is on, the sensor broadcasts the local event Alert to the Alarm
state. In the On substate of the state Alarm, the event triggers the transition from the Idle substate
to the Pending substate. When Pending becomes active, a warning sound alerts occupants to the
possible intrusion. If there is an accidental alarm, the occupants have a short time to disable the
security system. If not disabled within that time period, the system calls the police for help before
returning to the Idle mode.

Coordinate with Other Simulink Blocks

Stateflow charts can also use events to communicate with other blocks in a Simulink® model.

2-31
2 Stateflow Fundamentals

Output Events

An output event is an event that occurs in a Stateflow chart but is visible in Simulink blocks outside
the chart. This type of event enables a chart to notify other blocks in a model about events that occur
in the chart. For instance, in this example, the output events Sound and call_police drive external
blocks that handle the warning sound and the call to the police. The chart broadcasts these events
when the local event Alert triggers the transition to the Pending substate of the state Alarm. In
particular, in the Pending substate, the entry action broadcasts the Sound event. Similarly, the
condition action on the transition from Pending to Idle broadcasts the call_police event. In each
case, the action that broadcasts the output event uses the send operator with the name of the event:

send(eventName)

Each output event maps to an output port on the chart. Depending on the configuration, the
corresponding signal can control a Triggered Subsystem or a Function-Call Subsystem. To configure
an output event:

1 In the Modeling tab, under Design Data, select Symbols Pane and Property Inspector.
2 In the Symbols pane, select the output event.
3 In the Property Inspector, set Trigger to one of these options:

• Either edge — The output event broadcast causes the outgoing signal to toggle between zero
and one.
• Function call — The output event broadcast causes a Simulink function-call event.

In this example, the output events use edge triggers to activate a pair of latch subsystems in the
Simulink model. When each latch detects a change of value in its input signal, it briefly outputs a
value of one before returning to an output of zero.

Input Events

An input event is an event that occurs in a Simulink block but is visible in a Stateflow chart. This type
of event enables other Simulink blocks, including other Stateflow charts, to notify a specific chart of
events that occur outside it. For instance, in this example, the input event sl_call controls the
timing of the motion detector debouncer and the short delay before the call to the police. In each
instance, the event occurs inside a call to the temporal operator after, which triggers a transition
after the chart receives the event some number of times.

An external Simulink block sends an input event through a signal connected to the trigger port on the
Stateflow chart. Depending on the configuration, an input event results from a change in signal value
or through a function call from a Simulink block. To configure an input event:

1 In the Modeling tab, under Design Data, select Symbols Pane and Property Inspector.
2 In the Symbols pane, select the input event.
3 In the Property Inspector, set Trigger to one of these options:

• Rising — The chart activates when the input signal changes from either zero or a negative value
to a positive value.
• Falling — The chart activates when the input signal changes from a positive value to either zero
or a negative value.
• Either — The chart activates when the input signal crosses zero as it changes in either direction.

2-32
 Synchronize Parallel States by Broadcasting Events

• Function call — The chart activates with a function call from a Simulink block.

In this example, a Simulink Function-Call Generator block controls the timing of the security system
by triggering the input event sl_call through periodic function calls.

Explore the Example

In this example, the Stateflow chart has inputs from several Manual Switch blocks and outputs to a
pair of latch subsystems that connect to Display blocks. During simulation, you can:

• Enable the alarm and sensor subsystems and trigger intrusion detections by clicking the Switch
blocks.
• Watch the chart animation highlight the various active states in the chart.
• View the output signals in the Scope block and in the Simulation Data Inspector.

For example, suppose that you switch the alarm and sensor subsystems on, switch the sensor triggers
off, and start the simulation. During the simulation, you perform these actions:

1 At time seconds, you trigger the door sensor. The alarm begins to sound (Sound = 1) so
you immediately disable the alarm system. You switch the door sensor trigger off and turn the
alarm back on.
2 At time seconds, you trigger the window sensor and the alarm begins to sound (Sound =
0). This time, you do not disable the alarm. At around time , the security system calls the
police (call_police = 1). The Sound and call_police signals continue to toggle between
zero and one every 80 seconds.
3 At time seconds, you disable the alarm. The Sound and call_police signals stop
toggling.

The Simulation Data Inspector shows the response of the Sound and call_police signals to your
actions.

2-33
2 Stateflow Fundamentals

See Also
send

More About
• “Model Synchronous Subsystems by Using Parallel Decomposition” on page 2-25
• “Activate a Stateflow Chart by Sending Input Events”

2-34
 Synchronize Parallel States by Broadcasting Events

• “Activate a Simulink Block by Sending Output Events”

2-35
2 Stateflow Fundamentals

Monitor Chart Activity by Using Active State Data

If your Stateflow® chart includes data that is correlated to the chart hierarchy, you can simplify your
design by using active state data. By enabling active state data, you can:

• Avoid manual data updates reflecting chart activity.

• Log and monitor chart activity in the Simulation Data Inspector.
• Use chart activity data to control other subsystems.
• Export chart activity data to other Simulink blocks.

For example, in this model of a traffic signal, the state that is active determines the value of the
output signal color. You can simplify the design of the chart by enabling active state data. In this
case, the Stateflow chart can provide the color of the traffic signal by tracking state activity, so you do
not have to explicitly update the value of color.

To enable active state data, select a state to monitor. Then, in the Property Inspector:

1. Select Create output for monitoring.

2. Select one of these activity types:

• Self activity — Boolean value that indicates whether the state is active
• Child activity — Enumerated value that indicates which child state is active
• Leaf state activity — Enumerated value that indicates which leaf state is active

3. Enter the Data name for the active state data symbol.

4. Optionally, for Child activity or Leaf state activity, enter the Enum name for the active
state data type.

By default, Stateflow charts report state activity as output data to the Simulink model. To change the
scope of an active state data symbol to local data, use the Symbols pane.

2-36
 Monitor Chart Activity by Using Active State Data

Model a Traffic Signal Controller

This example uses active state data to model the controller system for a pair of traffic lights.

Inside the Traffic Controller chart, two parallel subcharts manage the logic that controls the
traffic lights. The subcharts have an identical hierarchy that consists of three child states: Red,
Yellow, and Green. The output data Light1 and Light2 correspond to the active child states in the
subcharts. These signals:

• Determine the phase of the animated traffic lights.

• Contribute to the number of cars waiting at each light.
• Drive a Safety Assertion subsystem that verifies that the two traffic lights are never green
simultaneously.

To see the subcharts inside the Traffic Controller chart, click the arrow at the bottom-left corner of
the chart.

2-37
2 Stateflow Fundamentals

Each traffic controller cycles through its child states, from Red to Green to Yellow and back to Red.
Each state corresponds to a phase in the traffic light cycle. The output signals Light1 and Light2
indicate which state is active at any given time.

Red Light

The traffic light cycle begins when the Red state becomes active. After a short delay, the controller
checks for cars waiting at the intersection. If it detects at least one car, or if a fixed length of time
elapses, then the controller requests a green light by setting greenLightRequest to true. After
making the request, the controller remains in the Red state for a short length of time until it detects
that the other traffic signal is red. The controller then makes the transition to Green.

Green Light

2-38
 Monitor Chart Activity by Using Active State Data

When the Green state becomes active, the controller cancels its green light request by setting
greenLightRequest to false. The controller sets greenLightLocked to true to prevent the
other traffic signal from turning green. After a short delay, the controller checks for a green light
request from the other controller. If it receives a request, or if a fixed length of time elapses, then the
controller transitions to the Yellow state.

Yellow Light

When the Yellow state becomes inactive, the controller sets greenLightLocked to false, indicating
that the other traffic light can safely turn green. The controller remains in the Yellow state for a
fixed amount of time before transitioning to the Red state. The traffic light cycle then begins again.

Timing of Traffic Lights

Several parameters define the timing of the traffic light cycle. To change the timing of the traffic
lights, double-click the Traffic Controller chart and, in the dialog box, enter new values for
these parameters:

• REDDELAY — Length of time before the controller checks for cars at the intersection. This value is
also the minimum length of time before the traffic light can turn green after the controller
requests a green light. The default value is 6 seconds.
• MAXREDDELAY — Maximum length of time that the controller checks for cars before requesting a
green light. The default value is 360 seconds.
• GREENDELAY — Maximum length of time that the traffic light remains green. The default value is
180 seconds.
• MINGREENDELAY — Minimum length of time that the traffic light remains green. The default value
is 120 seconds.
• YELLOWDELAY — Length of time that the traffic light remains yellow. The default value is 15
seconds.

Explore the Example

1 Open the chart by clicking the arrow in the bottom-left corner.

2 In the Symbols pane, select greenLightRequested. Then, in the Property Inspector, under
Logging, select Log signal data.
3 Repeat the previous step for greenLightLocked, Light1, and Light2.
4 In the Simulation tab, click Run.
5 In the Simulation tab, under Review Results, click Data Inspector.
6 In the Simulation Data Inspector, display the logged signals in separate axes. The Boolean signals
greenLightRequested and greenLightLocked appear as numeric values of zero or one. The
state activity signals Light1 and Light2 are enumerated data with values of Green, Yellow,
Red, and None.

2-39
2 Stateflow Fundamentals

To trace the chart activity during the simulation, you can use the zoom and cursor buttons in the
Simulation Data Inspector. For example, these are the key moments during the first 300 seconds of
the simulation:

• — At the start of the simulation, both traffic lights are red. Light1 and Light2 are Red,
greenLightRequested is false, and greenLightLocked is false.
• — After 6 seconds, the default value of REDDELAY, there are cars waiting in both streets, so
both traffic lights request a green light. Light1 and Light2 are still Red,
greenLightRequested is true, and greenLightLocked is false.
• — After another 6 seconds, the default value of REDDELAY, Light 1 becomes green, cancels
the green light request, and sets greenLightLocked to true. Then, Light 2 requests a green
light. Light1 is Green, Light2 is Red, greenLightRequested becomes false and then true,
and greenLightLocked is true.

2-40
 Monitor Chart Activity by Using Active State Data

• — After 120 seconds, the default value of MINGREENDELAY, Light 1 turns yellow. Light1
is Yellow, Light2 is Red, greenLightRequested is true, and greenLightLocked is true.
• — After 15 seconds, the default value of YELLOWDELAY, Light 1 turns red and sets
greenLightLocked to false. Then, Light 2 turns green, cancels the green light request, and
sets greenLightLocked to true. Light1 is Red, Light2 is Green, greenLightRequested is
false, and greenLightLocked becomes false and then true.
• — After 6 seconds, the default value of REDDELAY, Light 1 requests a green light. Light1
is Red, Light2 is Green, greenLightRequested is true, and greenLightLocked is true.
• — After Light 2 is green for 120 seconds, the default value of MINGREENDELAY, Light 2
turns yellow. Light1 is Red, Light2 is Yellow, greenLightRequested is true, and
greenLightLocked is true.
• — After 15 seconds, the default value of YELLOWDELAY, Light 2 turns red and sets
greenLightLocked to false. Then, Light 1 turns green, cancels the green light request, and
sets greenLightLocked to true. Light1 is Green, Light2 is Red, greenLightRequested is
false, and greenLightLocked becomes false and then true.
• — After 6 seconds, the default value of REDDELAY, Light 2 requests a green light. Light1
is Green, Light2 is Red, greenLightRequested is true, and greenLightLocked is true.

The cycle repeats until the simulation ends at seconds.

See Also

More About
• “Create a Hierarchy to Manage System Complexity” on page 2-21
• “Monitor State Activity Through Active State Data”
• “Simplify Stateflow Charts by Incorporating Active State Output”
• “View State Activity by Using the Simulation Data Inspector”

2-41
2 Stateflow Fundamentals

Schedule Chart Actions by Using Temporal Logic

To define the behavior of a Stateflow chart in terms of simulation time, include temporal logic
operators in the state and transition actions of the chart. Temporal logic operators are built-in
functions that tell you the length of time that a state remains active or that a Boolean condition
remains true. With temporal logic, you can control the timing of:

• Transitions between states

• Function calls
• Changes in variable values

These are the most common operators for absolute-time temporal logic:

• after — after(n,sec) returns true if n seconds of simulation time have elapsed since the
state that contains the operator or the source state of the transition that contains the operator
became active. Otherwise, the operator returns false. This operator supports event-based
temporal logic and absolute-time temporal logic in seconds (sec), milliseconds (msec), and
microseconds (usec).
• elapsed — elapsed(sec) returns the number of seconds of simulation time that have elapsed
since the activation of the associated state.
• duration — duration(C) returns the number of seconds of simulation time that have elapsed
since the Boolean condition C becomes true.

Model a Bang-Bang Temperature Controller

This example uses temporal logic to model a bang-bang controller that regulates the internal
temperature of a boiler.

The example consists of a Stateflow chart and a Simulink® subsystem. The Bang-Bang Controller
chart compares the current boiler temperature to a reference set point and determines whether to
turn on the boiler. The Boiler Plant Model subsystem models the dynamics inside the boiler by
increasing or decreasing its temperature according to the status of the controller. Then, the chart
uses the boiler temperature for the next step in the simulation.

The Bang-Bang Controller chart uses the temporal logic operator after to:

2-42
 Schedule Chart Actions by Using Temporal Logic

• Regulate the timing of the bang-bang cycle as the boiler alternates between on and off.
• Control a status LED that flashes at different rates depending on the operating mode of the boiler.

The timers defining the behavior of the boiler and LED subsystems operate independently of one
another without blocking or disrupting the simulation of the controller.

Timing of Bang-Bang Cycle

The Bang-Bang Controller chart contains a pair of substates that represent the two operating
modes of the boiler, On and Off. The chart uses the active state output data boiler to indicate which
substate is active.

The conditions on the transitions between the On and Off substates define the behavior of the bang-
bang controller:

• On the first transition from On to Off, the condition after(20,sec) turns the boiler off after it is
on for 20 seconds.
• On the transition from Off to On, the condition after(40,sec)[cold()] turns the boiler on
when the graphical function cold() indicates that the boiler temperature is below the reference
set point for at least 40 seconds.
• On the second transition from On to Off, the trivial condition turns the boiler off when the
internal transition logic in the On state determines that the boiler temperature is at or above the
reference set point.

As a result of these transition actions, the timing of the bang-bang cycle depends on the current
temperature of the boiler. At the start of the simulation, when the boiler is cold, the controller spends
40 seconds in the Off state and 20 seconds in the On state. At time seconds, the temperature
of the boiler reaches the reference set point. From that point, the boiler has to compensate only for
the heat lost while in the Off state. The controller then spends 40 seconds in the Off state and 4
seconds in the On state.

2-43
2 Stateflow Fundamentals

Timing of Status LED

The Off state contains a substate Flash with a self-loop transition triggered by the action
after(5,sec). Because of this transition, when the Off state is active, the substate executes its
entry action and calls the graphical function flash_LED every 5 seconds. The function toggles the
value of the output symbol LED between 0 and 1.

2-44
 Schedule Chart Actions by Using Temporal Logic

The On state calls the graphical function flash_LED as a combined entry, during state action.
When the On state is active, this action calls the function at every time step of the simulation to
toggle the value of the output symbol LED between 0 and 2.

As a result, the timing of the status LED depends on the operating mode of the boiler. For example:

• From to seconds, the boiler is off and the LED signal alternates between 0 and 1
every 5 seconds.
• From to seconds, the boiler is on and the LED signal alternates between 0 and 2
every second.
• From to seconds, the boiler is off and the LED signal alternates between 0 and 1
every 5 seconds.

2-45
2 Stateflow Fundamentals

Explore the Example

Use additional temporal logic to investigate how the timing of the bang-bang cycle changes as the
temperature of the boiler approaches the reference set point.

1. Enter new state actions that call the elapsed and duration operators:

• In the On state, set Timer1 to be the length of time that the On state is active:

en,du,ex: Timer1 = elapsed(sec);

• In the Off state, set Timer2 to be the length of time that the boiler temperature is at or above the
reference set point:

en,du,ex: Timer2 = duration(temp>=reference);

2-46
 Schedule Chart Actions by Using Temporal Logic

2. In the Symbols pane, click Resolve Undefined Symbols. The Stateflow Editor resolves the
symbols Timer1 and Timer2 as output data.

3. Enable logging for Timer1 and Timer2. In the Symbols pane, select each symbol. Then, in the
Property Inspector, under Logging, select Log signal data.

4. In the Simulation tab, click Run.

5. In the Simulation tab, under Review Results, click Data Inspector.

6. In the Simulation Data Inspector, display the signals boiler and Timer1 in the same set of axes.
The plot shows that:

• The On phase of the bang-bang cycle typically lasts 20 seconds when the boiler is cold and 4
seconds when the boiler is warm.
• The first time that the boiler reaches the reference temperature, the cycle is interrupted
prematurely and the controller stays in the On state for only 18 seconds.
• When the boiler is warm, the first cycle is slightly shorter than the subsequent cycles, as the
controller stays in the On state for only 3 seconds.

7. In the Simulation Data Inspector, display the signals boiler and Timer2 in the same set of axes.
The plot shows that:

• Once the boiler is warm, it typically takes 9 seconds to cool in the Off phase of the bang-bang
cycle.
• The first time that the boiler reaches the reference temperature, it takes 19 seconds to cool, which
is more than twice as long as the other cycles.

2-47
2 Stateflow Fundamentals

The shorter cycle and longer cooling time are a consequence of the substate hierarchy inside the On
state. When the boiler reaches the reference temperature for the first time, the transition from HIGH
to NORM keeps the controller on for an extra time step, which results in a warmer-than-normal boiler.
In later cycles, the history junction causes the On phase to start with an active NORM substate. The
controller then turns off immediately after the boiler reaches the reference temperature, which
results in a cooler boiler.

See Also

More About
• “Define Chart Behavior by Using State Actions and Transition Labels” on page 2-17
• “Control Chart Execution by Using Temporal Logic”
• “Reuse Logic Patterns by Defining Graphical Functions”
• “Resume Prior Substate Activity by Using History Junctions”

2-48
 3

Additional Learning Tools

3 Additional Learning Tools

Stateflow Onramp
Free, self-paced, interactive Stateflow course

Description
Stateflow Onramp is a free, self-paced, interactive course that helps you get started with Stateflow.

After completing Stateflow Onramp, you will be able to use the Stateflow environment and build
Stateflow charts based on real-life examples.

Stateflow Onramp teaches you to:

• Model state machines with Stateflow charts.

• Use Stateflow symbols and data.
• Control chart execution with actions.
• Simulate Stateflow charts with Simulink.
• Create flow charts to model common logic patterns.
• Improve chart readability and reuse code with functions.
• Organize charts using hierarchy.

Stateflow Onramp uses tasks to teach concepts incrementally, such as through a real-life example
with a robotic vacuum. You receive automated assessments and feedback after submitting tasks. Your
progress is saved when you exit the course, so you can complete the course in multiple sessions.

Open the Stateflow Onramp

• Simulink Start Page: On the Learn tab, click the Launch button that appears when you pause on
Stateflow Onramp.

3-2
 Stateflow Onramp

• Stateflow chart: On the quick access toolbar, click Help > Learn Stateflow.
• MATLAB Command Window: Enter learning.simulink.launchOnramp("stateflow").

Note If you do not have a Stateflow license, you can take the course at Self-Paced Online Courses.

Version History
Introduced in R2019b

See Also
learning.simulink.launchOnramp

Topics
“Model Finite State Machines” on page 2-2
“Construct and Run a Stateflow Chart” on page 2-10
“Define Chart Behavior by Using State Actions and Transition Labels” on page 2-17
“Create a Hierarchy to Manage System Complexity” on page 2-21

3-3