Trajectory Planning

Trajectory Planning

Path and trajectory planning means the way that a

robot is moved from one location to another in a
controlled manner.

The sequence of movements for a controlled

movement between motion segment, in straight-line
motion or in sequential motions.

It requires the use of both kinematics and dynamics of

robots.
 Trajectory Planning

Goal: to generate the reference inputs to the

motion control system which ensures that the
manipulator executes the planned trajectory

torques
Trajectory Motion control Robot
planning system system

Position, velocity, acceleration

 Trajectory Planning
PATH VS. TRAJECTORY

Path: A sequence of robot configurations in a

particular order without regard to the timing of
these configurations.
Trajectory: It concerned about when each part of
the path must be attained, thus specifying timing.

Sequential robot movements in a path.

 TRAJECTORY PLANNING

Path description
Joint (end-effector)
path constraints(obstacles)
trajectories in terms
constraints imposed by of position, velocity
robot dynamics (smooth) and acceleration
(

Trajectory planning
algorithm
 TRAJECTORY PLANNING
Plan a path in Cartesian Space

Method 1 Use inverse kinematics for

Cartesian Space
Techniques
finding the corresponding path
for the joints in the joint space.
Without
obstacle
Use inverse kinematics for
finding the initial and final joint
Method 2 position in joint space. Joint Space
Techniques

Plan a path in Joint Space

 TRAJECTORY PLANNING
Method 1 Method 2
Computing the inverse
kinematics in real time, Determine the initial and
time consuming and robot final joint positions. The
efficiency may be affected. path for all joints can be
easily planned in real time.
When a specific end effector
trajectory is required, this No need for forward and
method is useful. E.g. in inverse kinematics
the case of arc welding, recalculation.
the electrode is required
to follow the seam
precisely.
 TRAJECTORY PLANNING
Cartesian space
We can track a shape
(for orientation : equivalent axes, Euler angles,
More expensive at run time
(after the path is calculated need joint angles
in a lot of points)
Discontinuity problems

Joint space
Easy to go through via points
(Solve inverse kinematics at all path points and plan)
No problems with singularities
Less calculations
Can not follow straight line
 TRAJECTORY PLANNING
Cartesian planning
difficulties :
A
Initial and Goal
Points are C
reachable.

Intermediate points B

(C) unreachable.
 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle

The robot trajectory can be planned using initial

positions, initial velocity , final position and final velocity
of the joints in joint space. The third-order (cubic)
polynomial function is chosen as the position trajectory of
the ith joint:

i (t ) ai 0 ai1t ai 2t ai 3t
2 3

ai 0 , ai1 , ai 2 , ai 3 are constant parameters to be determined.

 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle

Differentiate with respect to time t,

i (t ) ai1 2ai 2t 3ai 3t 2

we obtained the velocity trajectory of the

ith joint.
 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle
i (t ) ai1 2ai 2t 3ai 3t 2

Assume that the initial position, final position, initial

velocity and final velocity are known, where the initial
time is t0=0 and final time is tf,
i (0) ai 0
i (t f ) ai 0 ai1t f ai 2t 2f ai 3t 3f
i (0) ai1
i (t f ) ai1 2ai 2t f 3ai 3t f2
 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle
i (t ) ai1 2ai 2t 3ai 3t 2

ai 0 i (0)
a (0)
i1 i

3( i (t f ) i (0)) (2i (0) i (t f ))t f

Obtain this ai 2
equation t 2f
2( i (0) i (t f )) (i (0) i (t f ))t f
ai 3
t 3f
Four parameters of the cubic polynomial can be uniquely
determined by the four constraints: the initial position,
final position, initial velocity and final velocity.
 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle
i (t ) ai1 2ai 2t 3ai 3t 2

When t0 is not equal to 0,

ai 0 i (t 0)
a (t 0)
i1 i

3( i (t f ) i (t 0)) (2i (t 0) i (t f ))(t f t0 )

ai 2
(t f t0 ) 2
2( i (t0 ) i (t f )) (i (t0 ) i (t f ))(t f t0 )
ai 3
(t f t0 ) 3
 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle

Consider a single link robot manipulator with a

rotary joint. Design a cubic trajectory, which starts
from the initial angular position i (0) 10 and ends
at the final angular position i (2) 90 , with zero
initial velocity and zero final velocity.
 TRAJECTORY PLANNING
WITH POLYNOMIALS
No obstacle
i (0) 10 i (2) 90 i (0) 0 i (2) 0

i (t ) ai 0 ai1t ai 2t ai 3t
2 3 a0 10
a1 0
3(90 10)
a2 2
60
2
2(10 90)
a3 3
20
2
Final trajectory,
(t ) 10 60t 20t
2 3
 EXERCISE 1
The second joint of SCARA manipulator is required
to move from 2 = 30 to 150 in 5 seconds. Find
the cubic polynomial to generate the smooth
trajectory for the joint. Assume that the initial and
final velocity of the joint is zero.
 POLYNOMIALS TRAJECTORIES WITH
VIA POINTS
With obstacle
Some via points must be specified in order to avoid collisions.

A
P(tf)

P(0) C
 POLYNOMIALS TRAJECTORIES WITH
VIA POINTS
With obstacle

Two methods fro trajectory planning with via point in space:

1. Use single high-order polynomial. The order of
polynomial depends on the number of the constraints.
2. With m via points, the whole trajectory could be divided
into m+1 segments, and each segment can be modeled
by using cubic polynomials.
 POLYNOMIALS TRAJECTORIES WITH
With obstacle
VIA POINTS

Consider a single-link robot manipulator with a

rotary joint. Design a single polynomial
trajectory, which starts from the initial angular
position (0)=10, passes the via point (1)=5,
and then stops at the final angular position
(2)=50, with zero initial velocity and zero final
velocity.
 POLYNOMIALS TRAJECTORIES WITH
With obstacle
VIA POINTS

The initial angular position (0) 10,

initial velocity (0) 0,
passes the via point (1) 5,
the final angular position (2) 50,
final velocity (2) 0 .
 POLYNOMIALS TRAJECTORIES WITH
With obstacle VIA POINTS
The order of polynomial depends on the number of
constraint. We have 5 constraints, fourth order polynomial
with five parameters as the candidate of the trajectory
function:
(t ) a0 a1t a2t a3t a4t
2 3 4

Velocity,
(t ) a1 2a2t 3a3t 2 4a4t 3
Acceleration,
(t ) 2a2 6a3t 12a4t 2
 POLYNOMIALS TRAJECTORIES WITH
With obstacle VIA POINTS
Using the five constraints, the following equation were
obtained,
a0 10
a1 0
a0 2a1 4a2 8a3 16a4 50
a1 4a2 12a3 32a4 0
a0 a1 a2 a3 a4 5

Then, a0 10, a1 0, a2 70, a3 90, a4 25

 POLYNOMIALS TRAJECTORIES WITH
With obstacle VIA POINTS
Then, a0 10, a1 0, a2 70, a3 90, a4 25
Fourth order polynomial trajectory is formed,

(t ) 10 70t 2 90t 3 25t 4

Velocity,

(t ) 140t 270t 100t
2 3

Acceleration,

(t ) 140 540t 300t 2
 TRAJECTORY PLANNING
Previously, the velocity increase until peak and then
reduce until zero when the robot arm move
towards the final position.

There are cases where the robot arm are required

to move at a constant velocity. In that case, the
trajectory should be divided into three segments
i.e. acceleration segment, constant velocity
segment and deceleration segment.
 POLYNOMIALS TRAJECTORIES WITH
With obstacle
VIA POINTS

Consider a single-link robot manipulator with a

rotary joint. Design a trajectory with following
cubic segments: the 1st segment connects initial
angular position (0)=10 to the via point
(1)=5, and 2nd segment that connects via
point (1)=5 to the final angular position
(2)=50. Assume that initial velocity and final
velocity are zero. At via point, the trajectory
should have continuous velocity and
acceleration
 TRAJECTORY PLANNING
Two parabolics model to define acceleration and
decelaration, one linear model to define
constant velocity segment, it is called Linear
Segment With Two Parabolic Blend (LPSB).