B.Tech.

(2nd Year)
Control System
By
Dr. Manoj Kumar

Robot Control and Observer scheme

➢ The word robot finds its origins in robota which means work in Czech.

➢ robot was introduced by the Czech science fiction writer Karel Capek to name artificial ˇ
humanoids – biped robots – which helped human beings in physically difficult tasks.

➢ These machines may be roughly classified as follows:

Both, mobile robots and manipulators are key pieces

of the mosaic that constitutes robotics nowadays.
➢ Robotics – a term coined by the science fiction writer Isaac Asimov – is as such a rather recent
field in modern technology.

➢ The good understanding and development of robotics applications are conditioned to the good
knowledge of different disciplines.

➢ Among these, electrical engineering, mechanical engineering, industrial engineering, computer

science and applied mathematics.

➢ Hence, robotics incorporates a variety of fields among which is automatic control of robot
manipulators.

➢ According to the definition adopted by the International Federation of Robotics under standard
ISO/TR 8373, a robot manipulator is defined as follows:
➢ A robot manipulator – or simply, manipulator – is a mechanical articulated arm that is
constituted of links interconnected through hinges or joints that allow a relative movement
between two consecutive links.
➢ The movement of each joint may be prismatic, revolute or a combination of both.

➢ Under reasonable considerations, the number of joints of a manipulator determines also its
number of degrees of freedom (DOF ). Typically, a manipulator possesses 6 DOF, among which 3
determine the position of the end of the last link in the Cartesian space and 3 more specify its
orientation.
The variables q1, q2 and q3 are referred to as the joint positions of the
robot. Consequently, these positions denote under the definition of an
adequate reference frame, the positions (displacements) of the robot’s
joints which may be linear or angular.

For analytical purposes, considering an n-DOF

robot manipulator, the joint positions are
collected in the vector q
➢ Physically, the joint positions q are measured by sensors conveniently
located on the robot.

➢ The corresponding joint velocities q˙ := d/dt q may also be measured or

estimated from joint position evolution.

➢ To each joint corresponds an actuator which may be electromechanical,

pneumatic or hydraulic.

➢ The actuators have as objective to generate the forces or torques which

produce the movement of the links and consequently, the movement of the
robot as a whole.

➢ For analytical purposes these torques and forces are collected in the vector
τ, i.e.
 What Does “Control of Robots” Involve? the physical variables of the system whose behavior
As a general rule, control design may be divided roughly into is desired to control. These may be temperature,
the following steps: pressure, displacement, velocity, etc. These variables
• familiarization with the physical system under consideration; are commonly referred to as the system’s outputs. In
• modeling; addition to this, we must also clearly identify those
• control specifications. variables that are available and that have an
influence on the behavior of the system and more
particularly, on its outputs. These variables are
referred to as inputs and may correspond for
instance, to the opening of a valve, voltage, torque,
force, etc.
Robots moving freely in their workspace, i.e. without For robots such as the one depicted in
interacting with their environment (cf. Figure 1.1) as Figure 1.2 that have physical contact with
for instance robots used for painting, “pick and place”, their environment, e.g. to perform tasks
laser cutting, etc., the output y to be controlled, may involving polishing, deburring of materials,
correspond to the joint positions q and joint velocities high quality assembling, etc., the output y
q˙ or alternatively, to the position and orientation of may include the torques and forces f
the end-effector (also called end-tool). exerted by the end-tool over its
environment.
Figure 1.3 shows a manipulator holding a marked tray, and a camera which provides an image of the tray with marks. The
output y in this system may correspond to the coordinates associated to each of the marks with reference to a screen on
a monitor.
Figure 1.4 depicts a manipulator whose end-effector has a camera attached to capture the scenery of its environment. In
this case, the output y may correspond to the coordinates of the dots representing the marks on the screen and which
represent visible objects from the environment of the robot.
we conclude that the corresponding output y of a robot system
– involved in a specific class of tasks – may in general, be of the
form
On the other hand, the input variables, that is, those that may be modified to affect the evolution of the output, are
basically the torques and forces τ applied by the actuators over the robot’s joints.
the block-diagram corresponding to the case when the outputs
are the joint positions and velocities, that is,

while τ is the input. In this case notice that for robots with n joints one has,
in general, 2n outputs and n inputs
 Motion Control of Robot Manipulators

➢ The simplest way to specify the movement of a manipulator is the so-called “point-to-point” method.
This methodology consists in determining a series of points in the manipulator’s workspace, which the
end-effector is required to pass through.
➢ Thus, the position control problem consists in making the end-effector go to a specified point
regardless of the trajectory followed from its initial configuration.
A more general way to specify a robot’s
motion is via the so-called (continuous)
trajectory. In this case, a (continuous)
curve, or path in the state space and
parameterized in time, is available to
achieve a desired task.
Then, the motion control problem
consists in making the end-effector
follow this trajectory as closely as
possible.

control problem, is also referred to as trajectory

tracking control.
The specified trajectory is simply a point in the workspace (which may be translated under appropriate conditions into a
point in the joint space). The position control problem consists in driving the manipulator’s end-effector (resp. the joint
variables) to the desired position, regardless of the initial posture.
➢ Gravity compensation is a well-known technique in robot design to
achieve equilibrium throughout the range of motion and as a result
to reduce the loads on the actuators.

gravitational torques g(q).

Computed Control Torque
 Computed Control Torque
Given a set of vectorial bounded functions qd, q˙ d and q¨d referred to as desired joint
positions, velocities and accelerations we wish to find a vectorial function τ such that the
positions q, associated to the robot’s joint coordinates follow qd accurately.
 Computed Control Torque
the vector q˜˙ (t) = q˙ d(t) - q˙ (t) stands for the velocity error. The control objective is achieved if the manipulator’s
joint variables follow asymptotically the trajectory of the desired motion.
However, for
practical
purposes it is
desirable that the
controller does
not depend on
the joint
acceleration q¨
since
accelerometers
are usually highly
sensitive to
noise.
Computed Control Torque
 Various controls
➢ Position Control: Position control refers to the ability to precisely control the position of the robot's end
effector (the part of the robot that interacts with the environment). This control allows the robot to move
to specific points in space with accuracy. For example, imagine a robot arm that needs to pick up an object
from a table. Position control enables the robot to move its arm to the exact location of the object to
perform the grasping task.
➢ Velocity Control: Velocity control involves regulating the speed at which the robot moves. It allows the
robot to adjust its speed while executing a task. For instance, if the robot needs to move its end effector
quickly to a certain position, velocity control ensures that it reaches that position at the desired speed.
Similarly, if a slow and precise movement is required, velocity control can be used to reduce the speed
accordingly.
➢ Trajectory Control: Trajectory control relates to defining and controlling the path that the robot's end
effector follows during a movement. It involves specifying a desired trajectory or path in space for the
robot to follow. For example, if the robot needs to trace a circular path with its end effector, trajectory
control enables the robot to precisely move along that circular trajectory, ensuring smooth and accurate
motion.
➢ Force Control: Force control refers to the ability of the robot to sense and control the forces it applies
during interactions with objects or the environment. It allows the robot to adjust its force or pressure when
manipulating objects, ensuring delicate handling or applying the right amount of force when needed. For
instance, a robot with force control can grip an object without crushing it, or it can apply a consistent force
while sanding a surface.
 Various controls

In summary, position control enables precise positioning of the

robot's end effector, velocity control regulates the speed of
movement, trajectory control determines the path to be followed,
and force control allows the robot to sense and control the
forces applied during interactions. These control techniques
collectively enable robots to perform various tasks accurately, adapt
to different situations, and safely interact with the environment.
 Proportional Control
Proportional Control (P): Proportional control, also known as P control, is a control strategy that uses the
proportional relationship between the error signal and the control output. The error signal represents the
difference between the desired value (setpoint) and the actual value being controlled. In P control, the control
output is proportional to this error signal.

example to help illustrate the concept: Suppose we have a robot arm that needs to maintain a specific
joint angle. The error signal in this case would be the difference between the desired angle and the
current angle of the joint. Proportional control would then scale this error signal by a gain factor
(proportional gain) and use it to directly control the joint torque or velocity. The higher the proportional
gain, the stronger the control response to the error signal.
 Derivative Control

Derivative Control (D): Derivative control, also known as D control, takes into account the rate of change of
the error signal. It adds a corrective action proportional to the derivative of the error signal over time. The
derivative term helps anticipate the future behavior of the error and helps dampen any sudden changes or
oscillations.

let's consider a scenario where the robot arm needs to track a moving target. The error signal, in this case, is
the difference between the desired position and the actual position of the end effector. The derivative
control calculates the rate of change of this error signal and multiplies it by a gain factor (derivative gain).
This derivative term contributes to the control output and helps the system respond more quickly to
changes, thereby reducing overshoot and improving stability.
 Proportional-Derivative Control

Proportional and derivative control are often combined in a control system known as proportional-
derivative (PD) control. PD control is a widely used control strategy that provides a good balance
between responsiveness and stability. By adjusting the gain values for the proportional and
derivative terms, engineers can fine-tune the robot's control system to achieve the desired
performance for various applications.

Proportional and Derivative Control (PD Control): PD control, combines proportional control (P)
and derivative control (D). It uses the error signal, which is the difference between the desired
value and the actual value being controlled, and applies a control output that is proportional to
the error signal and its rate of change. PD control is effective in achieving accurate and stable
motion control in robots.
 Proportional-Derivative Control with gravity Compensation

Proportional and Derivative Control with Gravity Compensation: Proportional and derivative control
with gravity compensation extends the concept of PD control by incorporating gravity
compensation. Gravity compensation aims to counteract the effects of gravity on the robot's motion,
particularly in robotic manipulators.

When a robot arm moves against gravity, it requires additional torque or effort to overcome the
gravitational forces acting on it. In proportional and derivative control with gravity compensation, an
additional term is added to the control system to compensate for the gravitational forces.

The gravity compensation term estimates the gravitational forces acting on the robot arm based on its
configuration (joint angles) and applies an opposing torque or force to counterbalance it. By doing so,
the robot arm can move more efficiently, requiring less effort from the control system to overcome
gravity.

Including gravity compensation in the control system is especially crucial for precise and accurate
motion control, as it allows the control system to focus on compensating for other disturbances or
errors rather than solely combating gravity.
Proportional-Derivative Control with gravity Compensation

while PD control combines proportional and derivative

control to achieve accurate and stable motion control,
proportional and derivative control with gravity
compensation goes a step further by incorporating a
term that compensates for the effects of gravity on the
robot's motion. By accounting for gravity, the control
system can operate more efficiently and focus on other
control tasks.
 Sliding Mode Control

Sliding mode control is a robust control strategy used in robot control to achieve precise and
robust tracking of desired trajectories. It is particularly effective in handling uncertainties and
disturbances in the system dynamics. The main idea behind sliding mode control is to drive
the system's state onto a predefined sliding surface and then maintain it on that surface
throughout the motion.

example of a robotic manipulator tasked with tracking a desired

trajectory. The goal is to move the end effector of the manipulator
along a specified path accurately, even in the presence of
uncertainties or disturbances.
 Sliding Mode Control
Sliding Surface Design: First, a sliding surface is defined based on the desired trajectory. The sliding
surface is a hyperplane or a specific mathematical function that represents the desired tracking behavior. It
typically involves a combination of the system's state variables, such as positions, velocities, or errors.

Control Input Design: Next, a control input is designed to drive the system's state onto the sliding
surface. This control input is often discontinuous and exhibits a switching behavior. It consists of two main
components:

a. Reaching Control: The reaching control component generates a control input

that forces the system's state to converge towards the sliding surface. It aims to
bring the system onto the sliding surface from any initial state within a finite time.

b. Sliding Control: Once the system reaches the sliding surface, the sliding control
component takes over. It continuously applies a control input to keep the system's
state on the sliding surface. This control input is designed to counteract
disturbances, uncertainties, or other external factors affecting the system's behavior.
 Sliding Mode Control
Robustness and Stability: One of the key advantages of sliding mode control is its robustness to
uncertainties and disturbances. The discontinuous nature of the control input ensures that the
system remains on the sliding surface, even in the presence of disturbances or uncertainties in the
system dynamics. Sliding mode control provides a high degree of robustness and can tolerate
significant modeling errors or external disturbances.

By continuously sliding along the predefined surface, the system can track the desired
trajectory accurately and robustly. Sliding mode control achieves excellent performance
even when the system dynamics are not perfectly known or when disturbances affect
the system's behavior.

Sliding mode control is a powerful technique for achieving precise and robust control in
robotics, particularly in scenarios where uncertainties and disturbances are present. It
ensures that the system remains on a predefined sliding surface, allowing accurate tracking
and resilience to external factors.
 Adaptive Control

Adaptive control is a control strategy used in robot control that enables the
robot system to adjust its control parameters or dynamics in real-time
based on the changing conditions of the environment or the system itself. It
is designed to handle uncertainties, variations, and changes in the system
dynamics, allowing the robot to adapt and perform optimally.

The main idea behind adaptive control is to continuously estimate or identify

the unknown or time-varying parameters of the robot system and update
the control parameters accordingly. This adaptive process helps the control
system to maintain desired performance despite uncertainties and changes.
 Adaptive Control

1. Parameter Estimation or Identification: Adaptive control relies on estimating or identifying the unknown or
time-varying parameters of the robot system. This involves developing algorithms or techniques that use
sensor measurements and system response data to evaluate the parameters. By continuously updating the
parameter estimates, the control system adapts to changes in the system dynamics.

2. Parameter Update: Once the parameter estimates are obtained, the control parameters are updated based on
the estimation results. The control parameters include gains, coefficients, or other variables that define the
behavior of the control system. The parameter update process ensures that the control system adjusts its
parameters to account for changes in the system dynamics and optimize the control performance.

3. Feedback Control: Adaptive control works in conjunction with feedback control. The control system receives
feedback information from sensors or measurements, compares it with the desired state or trajectory, and
generates control signals accordingly. The adaptive component adjusts the control parameters based on the
estimated or identified parameters, enhancing the control system's ability to adapt to changing conditions.

4. Robustness and Performance: Adaptive control enhances the robustness of the control system by adapting
to uncertainties and changes. It allows the control system to maintain desired performance levels even when
there are variations in the system dynamics, external disturbances, or uncertainties in the environment. By
adapting the control parameters, the system can optimize its performance and achieve the desired control
objectives.
 Adaptive Control Uses

Adaptive control is particularly useful in situations where the robot operates in

uncertain or changing environments or when the system dynamics vary over time. It
enables the robot to adapt and respond to these variations, ensuring robust and optimal
control performance.

Overall, adaptive control in robot control facilitates the ability of the control system to
continuously estimate and adjust its parameters, enabling the robot to adapt and perform
effectively in the face of uncertainties and changes in the system dynamics.
 Observer based control
Observer-based control, also known as state observer control, is a control strategy used in robot control
that combines a control law with an observer. The observer is responsible for estimating the unmeasured or
inaccessible states of the robot system, while the control law utilizes these state estimates to generate
control signals.

The observer, also called a state estimator, uses available sensor measurements and knowledge of the
system dynamics to estimate the current state of the robot. The estimated state information is then used by
the control law to generate control signals that drive the system towards the desired behavior.
Observer-based control involves the following key components and observer schemes:

1. Control Law: The control law defines the relationship between the estimated state and the desired
control signal. It uses the estimated state information to generate control inputs that steer the
robot towards the desired behavior or trajectory. The control law can be designed using various
techniques, such as proportional-derivative (PD) control, optimal control, or adaptive control.

2. Observer Design: The observer is responsible for estimating the unmeasured or inaccessible states
of the robot system based on available sensor measurements. There are different observer
schemes available, and the choice depends on the specific characteristics of the system and the
available sensor information. The two commonly used observer schemes are:
a. Luenberger Observer: The Luenberger observer is a widely used observer
scheme that estimates the state by comparing the measured sensor data with
the predicted system response. It employs a mathematical model of the system
dynamics and uses the sensor measurements to iteratively update the state
estimates.
b. Kalman Filter: The Kalman filter is an optimal observer scheme that estimates
the state of a dynamic system. It considers both the system model and sensor
measurements to estimate the state with minimum estimation error. The
Kalman filter incorporates statistical information about the system dynamics
and the measurement noise to improve the accuracy of the state estimates.
Feedback Loop: Observer-based control operates in a closed-loop feedback system. The sensor
measurements are used as inputs to both the observer and the control law. The observer estimates
the system states based on these measurements, and the estimated states are then used by the
control law to generate control signals. The control signals drive the robot system, and the process
repeats in a feedback loop to continually update the state estimates and control actions.

Observer-based control provides several benefits, including the ability to estimate unmeasured states,
handle sensor limitations, and enhance the system's robustness to disturbances and uncertainties. It is
commonly used in robot control applications where only partial information about the system states is
available or when accurate state estimation is critical for control performance.

Overall, observer-based control combines state estimation through an observer with control law design
to achieve accurate and robust control of the robot system, making it a valuable technique in robot
control.
 Robust Control

Robust Control: Robust control in robot control refers to the design of control systems that can
handle uncertainties, disturbances, and variations in the system dynamics. The objective of robust
control is to ensure that the system performs well despite these uncertainties, providing stability and
desired performance over a wide range of operating conditions.

Robust control techniques typically involve incorporating uncertainty models or bounds into the control
system design. These models or bounds account for variations in system parameters, external
disturbances, or modeling errors. The control system is then designed to provide stability and
performance guarantees even in the presence of these uncertainties.

Robust control methods include H-infinity control, mu-synthesis, and sliding mode control, among
others. These techniques aim to optimize the control system's performance and robustness, allowing
the robot to operate reliably and accurately in the face of uncertainties.
 Optimal Control

Optimal control in robot control refers to the design of control systems that minimize a specific
performance criterion. The objective is to find control inputs that optimize a defined cost or objective
function, such as minimizing energy consumption, maximizing precision, or reducing tracking error.

Optimal control techniques typically involve formulating an optimization problem with the desired
performance criterion as the objective function and system constraints. The control system design then
aims to find control inputs that minimize or maximize this objective function while satisfying the system
constraints.

There are various methods used in optimal control, such as Linear Quadratic Regulator (LQR), Model
Predictive Control (MPC), and Dynamic Programming. These methods utilize mathematical optimization
techniques to find control inputs that provide the best performance according to the defined criterion.
Observer Schemes: Observer schemes, as discussed in the previous response, are methods used to
estimate the unmeasured states of the system. Observers play a crucial role in state estimation,
allowing the control system to utilize estimated states for generating control signals.

Observer schemes, such as the Luenberger observer and Kalman filter, are used in various control
strategies, including robust control and optimal control. These observer schemes estimate the
unmeasured states based on available sensor measurements and system dynamics models. The
estimated states are then utilized in control law design to generate control signals that drive the robot
system.

In summary, robust control focuses on handling uncertainties and disturbances in the system dynamics
to provide stable and reliable control performance. Optimal control, on the other hand, aims to
minimize a performance criterion through optimization techniques. Observer schemes play a vital role in
estimating unmeasured states for both robust control and optimal control strategies.
 Design Based on Acceleration
To design a robot control system based on acceleration, you can follow these steps:
1.Define the Control Objective: Determine the specific objective of your robot control system. For
example, you may want the robot to reach a certain position or follow a specific trajectory while
considering acceleration constraints.
2.Sensor Integration: Integrate sensors such as accelerometers or gyroscopes into your robot. These
sensors will provide real-time measurements of the robot's acceleration.
3.State Estimation: Use sensor data and filtering techniques such as Kalman filters or particle filters
to estimate the robot's current state, including position, velocity, and acceleration.
4.Control Algorithm: Choose a control algorithm that takes acceleration into account. One commonly
used algorithm is a PID (Proportional-Integral-Derivative) controller, which can be extended to
incorporate acceleration constraints. Alternatively, you can use more advanced control techniques like
model predictive control (MPC) or optimal control methods.
5.Acceleration Constraints: Define the maximum allowable acceleration for your robot based on its
physical capabilities and safety considerations. Incorporate these constraints into the control algorithm
to ensure that the robot operates within safe limits.
6.Trajectory Planning: If your control objective involves following a specific trajectory, design a
trajectory planner that generates acceleration-limited trajectories. This planner will take into account
the desired path, the robot's acceleration constraints, and other factors like obstacle avoidance.
7. Feedback Loop: Implement a feedback loop that continuously measures the robot's
acceleration, compares it with the desired acceleration, and adjusts the control inputs accordingly.
This feedback loop ensures that the robot dynamically adapts its behavior based on acceleration
measurements.

8. Implementation and Testing: Implement the control algorithm on the robot's hardware platform
and thoroughly test its performance. Iterate and refine the design as needed based on the results
of testing.
Remember that designing a control system based on acceleration can be complex, and it may
require knowledge of control theory, robotics, and programming. Consulting relevant literature,
resources, or seeking the assistance of experts in the field can be beneficial during the design
process.