ME 652 Mobile Robotics: Handout #2: Robot Motion
ME 652 Mobile Robotics: Handout #2: Robot Motion
ME 652 Mobile Robotics: Handout #2: Robot Motion
ENGINEERING
ME 652
Mobile Robotics
Handout #2:
Robot Motion
Spring, 2020
Jinwhan Kim
KAIST
MECHANICAL
ENGINEERING
Classification of Vehicles
(Robotic) Ground Vehicles Legged
Vehicles
Wheeled
Rotary-wing
Underwater
Robot Locomotion
• Mechanisms or methods that a mobile robot use to
move from place to place, which requires physical
interaction between the robot and its environment.
• Bio-inspired robot locomotion: walking, jumping,
running, swimming, flying, etc.
Robot Locomotion
• Human inventions: using rotary mechanisms for
locomotion (e.g., wheels and propellers)
– Energy efficient and mechanically simple
– Instead of worrying about balance, we can focus on
more important problems (e.g., navigation, planning).
Coordinate Systems
• Cartesian (rectangular)
– In 3D,
– In 2D,
• Cylindrical (polar)
– In 3D,
– In 2D,
• Spherical
– In 3D,
Cartesian Coordinates
• Coordinates x, y , z
z
F • Unit vectors i, j, k
k (in directions of increasing
coordinates) are constant
i j
• Position vector
r=xi+yj+zk
r
z • Vector components
F = Fx i+Fy j+Fz k
= (F∙i) i+ (F∙j) j+ (F∙k) k
y x
y Components same regardless
of location of vector
x
ME652 by Jinwhan Kim, KAIST ME Adapted from lecture notes by Prof. Dean Mook @ Virginia Tech 8
MECHANICAL
ENGINEERING
Cylindrical Coordinates
• Coordinates r, , z
F
• Unit vectors er, e, ez
ez
e (in directions of increasing
coordinates)
er
• Position vector
R R = r er + z ez
z • Vector components
F = Fr er+F e +Fz ez
ME652 by Jinwhan Kim, KAIST ME Adapted from lecture notes by Prof. Dean Mook @ Virginia Tech 9
MECHANICAL
ENGINEERING
Spherical Coordinates
• Coordinates r, ,
er
e • Unit vectors er, e, e
(in directions of increasing
F
r coordinates)
e
• Position vector
r
r = r er
• Vector components
F = Fr er+F e +F e
ME652 by Jinwhan Kim, KAIST ME Adapted from lecture notes by Prof. Dean Mook @ Virginia Tech 10
MECHANICAL
ENGINEERING
Wheeled Robots
• Unicycle
• Steered wheels
(bicycle, tricycle)
• Differential drive
• Omnidirectional
• Tracked locomotion
• Walking wheels
Degrees of Freedom
# of degrees of freedom
of a given dynamical system
# of independent equations of
constraints
Motion Constraints
• A constraint is a restriction on the freedom of
movement of a system of particles, which specifies a
relation between the system’s coordinates.
• Equations of motion for any dynamical system can be
reduced into a set of simultaneous, coupled, 2nd
order differential equations associated which come
from Newton’s 2nd law applied to each mass
component.
• Any restrictions on the possible motion of a system
can be expressed by equations called constraint
equations.
Counting DOFs
• For holonomic systems, the number of DOFs of a
system is equal to the number of coordinates to
completely specify the system’s configuration.
# of independent constraints
Examples
Types of Constraints
• Holonomic vs. nonholonomic
– Holonomic: if constraints can be expressed in the following
form using generalized coordinates:
Holonomic or Nonholonomic
• A constraint is said to be holonomic if it can be
expressed in the form
Otherwise it is nonholonomic.
• Suppose you have a constraint of the form
Unicycle Example
• Equation of motion
Degrees of Freedom
• If any motion constraints of a system are not
integrable, the system is nonholomic. Also, the
number of DOFs is further reduced by the number of
nonholonomic constraints.
# of holonomic # of nonholonomic
constraints constraints
Nonholonomic Constraints
• Nonholonomic constraints typically arise in
the following classes of systems.
– No-slip constraint
• Mobile robots, any systems with pure rolling condition
– Conservation of angular momentum
• Free floating multibody systems
– Underactuated mechanical systems
• Most types of vehicles
• Actuators with passive joints
cf., the second-order nonholonomic systems
ME652 by Jinwhan Kim, KAIST ME 25
MECHANICAL
ENGINEERING
Underactuated Systems
• A system whose dimension of the configuration
space exceeds that of the control input space.
• For a control affine system
Underactuated Vehicles
• In fact, underactuated vehicles are very common in
real life. (e.g., wheeled mobile robots, aerial vehicles,
surface ships, underwater vehicles, etc.)
• Even fully actuated vehicles can be turned into
underactuated vehicle under certain circumstances
such as system failures and damage.
– Reliability considerations (fault-tolerant systems)
• This underactuated mechanism results in distinctive
motion characteristics, which requires special care in
motion planning.
Lie Bracket
• Suppose , where is a smooth vector
field
-f g
– It is important that f
-g
xo
xd
Examples
• An actuator with passive joints
• A car-like robot
• Turning of an astronaut
ME652 by Jinwhan Kim, KAIST ME 30