Joints task#
A JointsTask is a task that can be used to define a target value for some joints.
Warning
While it is possible to explicitly set a joint value by using set_joint()
on the robot wrapper (see joints configurations),
it is not equivalent to using a joint task.
Using the joints task will allow to account for everything
that the solver is enforcing (torque limits, joint limits, joint velocity limits, other tasks such as loops
closing constraints etc…), while set_joint() will immediately update
the joint value in the robot state regardless of the solver.
Task initialization#
You can use the following code to initialize the joint task:
# Initializing a joints task as a soft task with weight 1.0
joints_task = solver.add_joints_task()
joints_task.configure("joints", "soft", 1.0)
Task update#
You can configure the task by passing a dict to set_joints():
# Setting the target for the joints task
joints_task.set_joints({
"joint1": 0.5,
"joint2": -0.2,
})
Note
Unit values will depend on the joint types, it will be radians for revolute joints and meters for prismatic joints.
To specify the target velocity and acceleration, you can call set_joints()
on individual joints:
# Setting target position, velocity, and acceleration
joints_task.set_joint("shoulder_pan_joint", np.sin(t), np.cos(t), -np.sin(t))
Example#
Here is a complete example using a UR5 robot tracking a target frame. Its target velocity is also set for better tracking.
UR5 controlled with joints
In this example, a fixed contact is used on the base of the UR5 robot, which is following a sinusoidal joint-space trajectory.