0% found this document useful (0 votes)
108 views

Generalized Wheel Model

The document discusses the kinematics of wheeled mobile robots. It covers topics like degrees of freedom, forward and inverse kinematics, and developing mathematical models to relate wheel velocities to robot motion. Equations are presented to relate wheel velocities to the robot's linear and angular velocities.

Uploaded by

panave3104
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
108 views

Generalized Wheel Model

The document discusses the kinematics of wheeled mobile robots. It covers topics like degrees of freedom, forward and inverse kinematics, and developing mathematical models to relate wheel velocities to robot motion. Equations are presented to relate wheel velocities to the robot's linear and angular velocities.

Uploaded by

panave3104
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

Mobile Robot Kinematics

Wheeled Mobile Robots (WMRs)


• WMRs, as the name implies, have the ability to move around
(on the ground surface) with the help of wheels.
• DoF is three for the ground/land-based wheeled mobile
robots. i.e., two translations and one orientation in a plane.
Kinematic Simulation of a Mobile Robot
• Kinematics is the study of (the mathematics) motion without considering
the forces/efforts that effect the motion.
• Kinematic simulation of a mobile robot involves modeling the motion of the
robot and predicting its behavior based on its kinematic properties.
• Kinematic simulations help in designing, controlling, and optimizing the
performance of a mobile robot.
Need of a Mathematical Model
• To understand the behavior of the system and design the system,
• To design suitable controllers, navigation systems and adjust their
performances,
• To predict or estimate the system parameters, to illustrate or mimic
or simulate the real system, etc.
Degree of freedom (DoF):
• Minimum number of variables (or number of
independent variables) to describe the system.
• For the land-based wheeled mobile robots
and water surface vehicles: degree of freedom
is three (two translations and one orientation
in a plane).
• For the aerial, space and underwater robots:
degree of freedom is six (three translations
and three orientations in space).
Forward Differential Kinematics of Mobile Robot
• Forward kinematics is the process of determining the position and
orientation of a robot's end-effector (or any specific point on the robot)
based on its joint angles and link lengths.
• For mobile robots, forward kinematics typically involves calculating the
robot's position and orientation in the workspace based on its wheel
velocities and geometric parameters.
• Simulating or analyzing the system in velocity level.
Forward Differential Kinematics of a Mobile Robot
x Mobile robot

v
u
𝒚ሶ = 𝒖 𝒔𝒊𝒏 𝝍ሶ + 𝒗 𝒄𝒐𝒔 𝝍 v 𝑣 cos 𝜓
r

𝑦𝑏 𝑥𝑏 𝝍
𝑣 cos 𝜓 ● 𝑣 sin 𝜓
B
𝑦ሶ
u
y 𝑢 sin 𝜓
𝑢 sin 𝜓
𝑢 cos 𝜓
𝑦𝑖
I: Inertial Coordinate of the Mobile Robot about a fixed point
I● B: Body fixed coordinate of the Mobile Robot
𝑥𝑖
x: Forward displacement of the mobile robot w.r.t. I
𝑣 sin 𝜓 y: Lateral displacement of the mobile robot w.r.t. I
𝝍 : Angular displacement of the mobile robot w.r.t. I
𝑢 cos 𝜓 u: Forward velocity of the mobile robot w.r.t. B
𝑥ሶ
v: Lateral velocity of the mobile robot w.r.t. B
𝑥ሶ = 𝑢 cos 𝜓 − 𝑣 sin 𝜓
r : Angular velocity of the mobile robot w.r.t. B
𝑥ሶ = 𝑢 cos 𝜓 − 𝑣 sin 𝜓
𝑦ሶ = 𝑢 𝑠𝑖𝑛 𝜓ሶ + 𝑣 𝑐𝑜𝑠 𝜓
𝜓ሶ = r

Rewriting in matrix form,


𝑥ሶ 𝑢 cos 𝜓 − 𝑣 sin 𝜓
𝑦ሶ = 𝑢 sin 𝜓 + 𝑣 cos 𝜓
𝜓ሶ 𝑟
Inertial fixed time derivative
of the generalized coordinate

Body fixed
𝑥ሶ cos 𝜓 −𝑠𝑖𝑛 𝜓 0 𝑢 Instantaneous
𝑦ሶ = sin 𝜓 cos 𝜓 0 𝑣 velocity
𝜓ሶ 0 0 1 𝑟

𝜂ሶ = 𝐽 𝜓 𝜍
It describes the relation between the velocity input commands 𝜍 and the derivatives of generalized coordinates 𝜂.ሶ

𝐽 𝜓 is the Jacobian (or velocity transformation) matrix.


The known mobile robot Forward kinematic model, as:

𝜂ሶ = 𝐽 𝜓 𝜍
Based on wheel configuration

Where:

W= Wheel Configuration Matrix which depends on the type of wheel


𝜔 = Vector of angular velocity of the wheel
𝜂ሶ = is the vector of time derivative of generalized coordinates
𝐽 𝜓 = Jacobian matrix which maps the input velocity commands to
the derivatives of the generalized coordinates
𝜍 = Vector of velocity input commands
Inverse Differential Kinematics of a Mobile Robot

• Inverse kinematics refers to the process of determining the joint


parameters or velocities required to achieve a desired end-effector
position and orientation.
• For mobile robots, inverse kinematics involves calculating the required
wheel velocities to achieve a desired linear and angular velocity for the
given position trajectory.
−1
𝜍=𝐽 𝜓 𝜂ሶ

Controlling the system in velocity level.


A Generalized Wheel Model
Let us consider a mecanum wheel mobile robot configuration with four Mecanum
wheels

𝑦𝑏
𝑥𝑏
B

𝑦𝑖

I 𝑥𝑖
Let us consider one of the ith wheel
mecanum wheel of the
mobile robot configuration

𝑦𝑏

● 𝑥𝑏
B

I: Inertial Coordinate of the Mobile Robot about a fixed point


B: Body fixed coordinate of the Mobile Robot
x: Forward displacement of the mobile robot w.r.t. I
𝑦𝑖 y: Lateral displacement of the mobile robot w.r.t. I
𝝍 : Angular displacement of the mobile robot w.r.t. I
I 𝑥𝑖 u: Forward velocity of the mobile robot w.r.t. B
v: Lateral velocity of the mobile robot w.r.t. B
r : Angular velocity of the mobile robot w.r.t. B
V 𝑠𝑙𝑖𝑑𝑒 V drive

𝑦𝑏
𝑥𝑏
B● th
V drive - Tangential velocity of the i wheel
th
𝜔𝑖 - Angular velocity of the i wheel.
th
𝑎𝑖 - Radius of the i wheel
V drive = 𝝎𝒊 × 𝒂𝒊

𝑦𝑖
V 𝑠𝑙𝑖𝑑𝑒 - Tangential velocity of the Passive roller
𝛽𝑖ሶ - Angular velocity of the passive roller
I 𝑥𝑖
𝜌𝑖 - Radius of the passive roller
V 𝒔𝒍𝒊𝒅𝒆 = 𝜷𝒊ሶ × 𝝆𝒊
V drive = 𝜔𝑖 𝑎𝑖 From the diagram we can write:
𝛽𝑖ሶ 𝜌𝑖 𝑐𝑜𝑠 ∅𝑖
Lateral velocity at Ci, 𝑦𝑐𝑖ሶ = 𝛽𝑖ሶ 𝜌𝑖 cos ∅𝑖
V 𝑠𝑙𝑖𝑑𝑒 = 𝛽𝑖ሶ 𝜌𝑖 ∅𝑖
𝑦𝐶𝑖 𝑦𝑐𝑖ሶ
𝑥𝐶𝑖 ⇒ 𝛽𝑖ሶ 𝜌𝑖 =
● cos ∅𝑖
𝑪𝒊
𝑥𝑐𝑖ሶ = 𝜔𝑖 𝑎𝑖 − 𝛽𝑖ሶ 𝜌𝑖 sin ∅𝑖
𝑦𝑐𝑖ሶ
𝛽𝑖ሶ 𝜌𝑖 sin ∅𝑖 𝑥𝑐𝑖ሶ = 𝜔𝑖 𝑎𝑖 − sin ∅𝑖
cos ∅𝑖
th
𝑪𝒊 - wheel frame center of the i wheel
th
𝑥𝑐𝑖ሶ = 𝜔𝑖 𝑎𝑖 − 𝑦𝑐𝑖ሶ tan ∅𝑖
V drive - Tangential velocity of the I wheel
𝜔𝑖 - Angular velocity of the i
th
wheel. ⇒ 𝜔𝑖 𝑎𝑖 = 𝑥𝑐𝑖ሶ + 𝑦𝑐𝑖ሶ tan ∅𝑖
th
𝑎𝑖 - Radius of the i wheel. Rewriting the above
V 𝑠𝑙𝑖𝑑𝑒 - Tangential velocity of the Passive roller
1
𝛽𝑖ሶ - Angular velocity of the passive roller ⇒ 𝜔𝑖 = (𝑥𝑐𝑖ሶ + 𝑦𝑐𝑖ሶ tan ∅𝑖 )
𝑎𝑖
𝜌𝑖 - Radius of the passive roller
Rearranging in matrix form
∅𝑖 - Angle between roller axis to the Xci axis.
𝑦𝑐𝑖ሶ - Lateral Velocity 1 1 𝑥𝑐𝑖ሶ
𝜔𝑖 = tan ∅𝑖 --- (1)
𝑥𝑐𝑖ሶ - Longitudinal velocity 𝑎𝑖 𝑎𝑖 𝑦𝑐𝑖ሶ
Let us assume that for an arbitrary case Linear velocity vector of 𝑐𝑖 with respect to the frame of 𝑐𝑖
is given by :
𝜃𝐵𝑖 is the angle between 𝑥𝐶𝑖 and 𝑥𝑏 𝑥𝑐𝑖ሶ
𝐶𝑖
.𝑣𝐶𝑖 =
𝑦𝑐𝑖ሶ
𝑦𝑐𝑖ሶ But we need Linear velocity vector of 𝑐𝑖 with respect to the frame B
𝜃𝐵𝑖 𝑥𝑐𝑖ሶ
𝑥𝐶𝑖 𝐵 𝐵 𝑥ሶ 𝑐𝑖
𝑦𝐶𝑖 𝜃𝐵𝑖 .𝑣𝐶𝑖 = 𝑐𝑖 𝑅(𝜃𝐵𝑖 ) 𝑦ሶ 𝑐𝑖

𝑪𝒊
Velocity component along 𝑥𝐵 , 𝑥ሶ 𝑐𝑖 cos 𝜃𝐵𝑖 − 𝑦ሶ 𝑐𝑖 sin 𝜃𝐵𝑖
Velocity component along 𝑦𝐵 , 𝑥ሶ 𝑐𝑖 sin 𝜃𝐵𝑖 + 𝑦ሶ 𝑐𝑖 cos 𝜃𝐵𝑖
} ---(2)
𝑦𝑏


𝐵 𝑥𝑏
𝐵𝑣 cos 𝜃𝐵𝑖 − sin 𝜃𝐵𝑖 𝑥ሶ 𝑐𝑖
. 𝐶𝑖 =
𝑥𝑐𝑖ሶ - Longitudinal velocity sin 𝜃𝐵𝑖 cos 𝜃𝐵𝑖 𝑦ሶ 𝑐𝑖
𝑦𝑐𝑖ሶ - Lateral Velocity
𝜃𝐵𝑖 - Angle between the vehicle frame (B) to the wheel frame (ci ).
𝐶𝑖 𝑣 - Linear velocity vector of 𝑐 with respect to the frame of 𝑐
. 𝐶𝑖 𝑖 𝑖
𝐵
.𝑣𝐶𝑖 - Linear velocity vector of 𝑐𝑖 with respect to the frame B
𝐵
𝑐𝑖 𝑅(𝜃𝐵𝑖 ) –Rotation Matrix
The frame B and Ci are part of a single body, thus if B is rotated Ci also will be rotated
If u, v, r are the instantaneous velocities at B, then equivalent velocity at 𝐶𝑖 can be given by
From Fig. we can write,
𝑣
Velocity component along 𝑥𝑏 = 𝑢 − 𝑟𝑑𝑦𝑖
𝑟𝑑𝑥𝑖
Velocity component along 𝑦𝑏 =𝑣 + 𝑟𝑑𝑥𝑖
} --- (3)
𝑥𝑐𝑖
𝑟𝑑𝑦𝑖 ●𝑐
𝜃𝐵𝑖
𝑖 𝑢
1 0 −𝑑𝑦𝑖 𝑢
𝐵𝑣 = 𝑣
. 𝐶𝑖
0 1 𝑑𝑥𝑖 𝑟
𝑑𝑦𝑖

𝑣 Equating 2 & 3,
𝑦𝑏 r 𝑢 − 𝑟𝑑𝑦𝑖 = 𝑥ሶ 𝑐𝑖 cos 𝜃𝐵𝑖 − 𝑦ሶ 𝑐𝑖 sin 𝜃𝐵𝑖
𝐵

𝑥𝑏
𝑢
𝑣 + 𝑟𝑑𝑥𝑖 = 𝑥ሶ 𝑐𝑖 sin 𝜃𝐵𝑖 + 𝑦ሶ 𝑐𝑖 cos 𝜃𝐵𝑖
} --- (4)
𝑑𝑥𝑖

dxi and dyi are the position coordinates of ci with reference to B.


Writing above in matrix form we get

1 0 −𝑑𝑦𝑖 𝑢
cos 𝜃𝐵𝑖 − sin 𝜃𝐵𝑖 𝑥ሶ 𝑐𝑖
0 1 𝑑𝑥𝑖
𝑣 =
𝑟
sin 𝜃𝐵𝑖 cos 𝜃𝐵𝑖 𝑦ሶ 𝑐𝑖 } --- (5)
By matrix manipulation we get,
−1
1 0 −𝑑𝑦𝑖 𝑢
𝑥ሶ 𝑐𝑖 cos 𝜃𝐵𝑖 − sin 𝜃𝐵𝑖
= 𝑣
𝑦ሶ 𝑐𝑖 sin 𝜃𝐵𝑖 cos 𝜃𝐵𝑖 0 1 𝑑𝑥𝑖 𝑟
1 0 −𝑑𝑦𝑖 𝑢
𝑥ሶ 𝑐𝑖 cos 𝜃𝐵𝑖 sin 𝜃𝐵𝑖
= 𝑣 --- (6)
𝑦ሶ 𝑐𝑖 −sin 𝜃𝐵𝑖 cos 𝜃𝐵𝑖 0 1 𝑑𝑥𝑖 𝑟
Substituting eqn.(6) in eqn.(1),
1 1 𝑢
cos 𝜃𝐵𝑖 sin 𝜃𝐵𝑖 1 0 −𝑑𝑦𝑖
𝜔𝑖 = tan ∅𝑖 𝑣 --- (7)
𝑎𝑖 𝑎𝑖 −sin 𝜃𝐵𝑖 cos 𝜃𝐵𝑖 0 1 𝑑𝑥𝑖
𝑟
We have Forward kinematic model of the Mobile Robot as,
𝑥ሶ cos 𝜓 −𝑠𝑖𝑛 𝜓 0 𝑢
𝑦ሶ = sin 𝜓 cos 𝜓 0 𝑣
𝜓ሶ 0 0 1 𝑟

By Matrix operation
−1 𝑥ሶ
𝑢 cos 𝜓 −𝑠𝑖𝑛 𝜓 0
𝑣 = sin 𝜓 cos 𝜓 0 𝑦ሶ
𝑟 0 0 1 𝜓ሶ

𝑢 cos 𝜓 𝑠𝑖𝑛 𝜓 0 𝑥ሶ
𝑣 = − sin 𝜓 cos 𝜓 0 𝑦ሶ
𝑟 0 0 1 𝜓ሶ
Substituting and simplifying eqn.(7) we get,
cos 𝜓 𝑠𝑖𝑛 𝜓 0 𝑥ሶ
1 1 cos 𝜃𝐵𝑖 sin 𝜃𝐵𝑖 1 0 −𝑑𝑦𝑖
𝜔𝑖 = tan ∅𝑖 − sin 𝜓 cos 𝜓 0 𝑦ሶ
𝑎𝑖 𝑎𝑖 −sin 𝜃𝐵𝑖 cos 𝜃𝐵𝑖 0 1 𝑑𝑥𝑖
0 0 1 𝜓ሶ

You might also like