ED 5315
INTRODUCTION TO FIELD AND SERVICE ROBOTICS
Instructor: Bijo Sebastian
LOCOMOTION
Locomotion - Means by which to move unbounded though the environment
Different approaches found in nature: Roll, walk/run, swim, fly, etc.
+ Successful in wide variety of harsh environments
- High mechanical complexity
Biped walking ~ rolling polygon
Step length = Polygon side length
Smaller step rolling of a circle
2
LEGGED LOCOMOTION
Key advantages:
- Adaptability and maneuverability in rough terrain
- Traverse obstacles
- Reduced environmental impact
Disadvantages:
- Power consumption: Oscillating motion energy loss
Dynamically stable passive walking is possible [Link]
- Mechanical design and control complexity : at least
three points of contact for static stability
- Legged locomotion can be considered as the compliment
of manipulation
Statically stable walking with six legs 3
Source: [Link]
LEGGED LOCOMOTION
- Single legged robot
- Bipeds
- Quadrupeds
- Hexapods
Railbert hopper, Leg Lab MIT
Legged robots need more research to
perform closely to their biological
equivalents
Atlas, Boston Dynamics
MX-Phoenix hexapod LittleDog, Boston Dynamics
4
WHEELED LOCOMOTION
3 or more wheels in contact (static) stable locomotion Most popular robot locomotion mechanism
Compared to legged locomotion: Focus shifts to traction, maneuverability, and control
Wheel designs
- Standard wheel: 2 DOF
- Castor wheel: 2 DOF
- Swedish wheel/Mecanum wheel: 3DOF
- Spherical wheel:3 DOF
5
WHEELED LOCOMOTION
3 or more wheels in contact stable locomotion Most popular robot locomotion mechanism
Compared to legged locomotion: Focus shifts to traction, stability, maneuverability, and control
Wheel designs
- Standard wheel: 2 DOF
- Castor wheel: 2 DOF
- Swedish wheel/Mecanum wheel: 3 DOF
- Spherical wheel:3 DOF
6
WHEELED LOCOMOTION
3 or more wheels in contact stable locomotion Most popular robot locomotion mechanism
Compared to legged locomotion: Focus shifts to traction, stability, maneuverability, and control
Wheel designs
- Standard wheel: 2 DOF Swedish 90°
- Castor wheel: 2 DOF
- Swedish wheel/Mecanum wheel: 3DOF
- Spherical wheel:3 DOF
Swedish 45°
7
WHEELED LOCOMOTION
3 or more wheels in contact stable locomotion Most popular robot locomotion mechanism
Compared to legged locomotion: Focus shifts to traction, stability, maneuverability, and control
Wheel designs
- Standard wheel: 2 DOF
- Castor wheel: 2 DOF
- Swedish wheel/Mecanum wheel: 3DOF
- Spherical wheel:3 DOF
[Link] 8
WHEELED LOCOMOTION
3 or more wheels in contact stable locomotion Most popular robot locomotion mechanism
Compared to legged locomotion: Focus shifts to traction, stability, maneuverability, and control
Wheel designs
- Standard wheel: 2 DOF
- Castor wheel: 2 DOF
- Swedish wheel/Meccannum wheel: 3DOF
- Spherical wheel:3 DOF
[Link] 9
WHEEL CONFIGURATION
Deciding factors for wheel type and wheel geometry:
• Maneuverability
Best with omnidirectional wheels (Swedish or spherical)
Least for standard wheels on a steering system (Ackermann)
• Controllability
Omnidirectional least (slippage, complex design)
Better for less maneuverable systems
• Stability
Minimum one (dynamic stability), static stability three
More than 3 wheel + non-uniform terrain
Suspension system (Irrespective of the type of wheel)
10
WHEEL CONFIGURATION
Some notable configurations:
• Bicycle model:
One steering wheel (front)
One traction wheel (back)
• Differential drive robot:
Two powered wheels
one castor wheel
11
WHEEL CONFIGURATION
Some notable configurations:
• Ackerman model:
Two motorized and steered wheels in front
Two free wheels in rear
Steering angle for two front wheels will have to be different to avoid skidding
• Synchro drive:
??
12
WHEEL CONFIGURATION
Additional reading:
• Chapter 2 of “Introduction to Autonomous
Mobile robots” book, especially Section 2.3
13
WHEEL CONFIGURATION
• Synchro drive:
All wheels are actuated synchronously by one motor
All wheels are steered synchronously by a single motor
Question: Can we control orientation of platform?
14
WHEEL CONFIGURATION
• Synchro drive:
All wheels are actuated synchronously by one motor
All wheels are steered synchronously by a single motor
Not possible to control orientation of platform
15
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Mobility:
Every wheel adds constraint of no lateral slip on the robot
platform, provided the wheel cannot turn freely
Can be visualized through the concept of robot’s instantaneous
center of rotation (ICR)
ICR can be located anywhere
Degree of mobility: 3
16
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Mobility:
Every wheel adds constraint of no lateral slip on the robot
platform, provided the wheel cannot turn freely
Can be visualized through the concept of robot’s instantaneous
center of rotation (ICR)
ICR can be located anywhere
ICR can be located anywhere on
Degree of mobility: 3
the line (variable arc motion)
Degree of mobility: 2
17
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Mobility:
Every wheel adds constraint of no lateral slip on the robot
platform, provided the wheel cannot turn freely
Can be visualized through the concept of robot’s instantaneous
center of rotation (ICR)
ICR can be located anywhere
ICR can be located anywhere on ICR is a fixed point if steering
Degree of mobility: 3
the line (variable arc motion) held fixed (fixed arc motion)
Degree of mobility: 2 Degree of mobility: 1
(Same for a bicycle)
18
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Steerability:
Every independently controllable steerable wheel adds a degree of
steerability
No steerable wheels
Degree of steerability: 0
19
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Steerability:
Every independently controllable steerable wheel adds a degree of
steerability
Two mutually dependent
No steerable wheels steerable wheels
Degree of steerability: 0 Degree of steerability: 1
20
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Steerability:
Every independently controllable steerable wheel adds a degree of
steerability
Two mutually dependent Two mutually independent
No steerable wheels steerable wheels steerable wheels
Degree of steerability: 0 Degree of steerability: 1 Degree of steerability: 2
21
MANEUVERABILITY OF MOBILE ROBOT
• Degree of Maneuverability:
The overall degrees of freedom that a robot can manipulate
Degree of maneuverability = Degree of mobility+ Degree of steerability
𝛿 =𝛿 +𝛿
Degree of maneuverability 𝑖𝑠 𝑛𝑜𝑡 𝑎𝑙𝑤𝑎𝑦𝑠 𝑒𝑞𝑢𝑎𝑙 𝑡𝑜 Degree of freedom
22
KINEMATIC MODEL OF A ROBOT
Mathematical models are useful in predicting the behaviour (future states) of a system based on previous state
and input.
Application in control, planning, state estimation (localization) etc..
0,1,1,2,3,_,_,_.....
Kinematics: Study of motion of points, bodies (objects), and systems of bodies (groups of objects) without
considering the forces that cause them to move.
Kinematic model of the robot can be derived from the constraints on how a mobile robot moves in its
workspace
23
KINEMATIC MODEL OF A ROBOT
Kinematics model of the robot places constraints on how the mobile robot moves in its workspace
Unicycle:
[Link]
24
KINEMATIC MODEL OF A ROBOT
Y Y’
Representing the mobile robot position:
X’
3 DOF system: x, y position and θ orientation
𝜃
𝑦
(Any additional internal degrees of freedom or flexibility P
within the robot is ignored)
Pose of the robot (in global frame), O 𝑥 X
𝑥
𝑋= 𝑦
𝜃
25
KINEMATIC MODEL OF A ROBOT
Y
Two inputs 𝑣, 𝜔 and three 3 states 𝑥, 𝑦, 𝜃 𝜔
𝜃
𝑥̇ = 𝑣 cos 𝜃 𝑦
𝑥̇
𝑋̇ = 𝑦̇ 𝑦̇ = 𝑣 𝑠𝑖𝑛 𝜃
𝜃̇
𝜃̇ = 𝜔
O 𝑥 X
New state is given by;
𝑋 = 𝑋 + 𝑋̇Δ𝑡
26
HOLONOMY
• Holonomic kinematic constraint: Can be expressed solely as a function of state variables
• Non-Holonomic constraint requires a differential relationship. Integrating the relationship will not give a
meaningful constraint, hence non-integrable
• Holonomic robot has zero non-holonomic constraints either no constraints at all or only holonomic
constraints, e.g.: omnidirectional robot, bicycle with locked steering wheel
• Omnidirectional robots are holonomic
• Differential drive robots, Ackerman steering or Bicycle model robots are all non-holonomic
• If application requires maneuvering around obstacles without affecting orientation or tracking an object
while following a random trajectory, then you need to use holonomic robot
27
LOCOMOTION
Extra reading:
• Chapter 2 of “Introduction to Autonomous Mobile robots” book
28
