Adaptive Cartesian Compliance Controller

Extension of the cartesian compliance controller to work in an adaptive scenario.

Example Configuration

Below is an example controller_manager.yaml for a controller specific configuration. Also see the simulation config for further information.

controller_manager:
  ros__parameters:
    update_rate: 500  # Hz

    cartesian_adaptive_compliance_controller:
      type: cartesian_adaptive_compliance_controller/CartesianAdaptiveComplianceController

    # More controller instances here
    # ...

cartesian_adaptive_compliance_controller:
  ros__parameters:
    end_effector_link: "tool0"
    robot_base_link: "base_link"
    ft_sensor_ref_link: "sensor_link"
    compliance_ref_link: "tool0"
    joints:
      - joint1
      - joint2
      - joint3
      - joint4
      - joint5
      - joint6

    command_interfaces:
      - position

    stiffness:  # minimum stiff. values w.r.t. compliance_ref_link
        trans_x: 500
        trans_y: 500
        trans_z: 500
        rot_x: 20
        rot_y: 20
        rot_z: 20

    solver:
        error_scale: 0.5
        iterations: 1

    pd_gains:
        trans_x: {p: 0.05, d: 0.005}
        trans_y: {p: 0.05, d: 0.005}
        trans_z: {p: 0.05, d: 0.005}
        rot_x: {p: 1.5}
        rot_y: {p: 1.5}
        rot_z: {p: 1.5}

    # Parameters of the QP and adaptive controller
    tank:
        initial_state: 1.0

    qp:
        Fmin:

    Q_weights:
        - 3200.0
        - 3200.0
        - 3200.0

    R_weights:
        - 0.01
        - 0.01
        - 0.01

    F_min:
        - -15.0
        - -15.0
        - -15.0

    F_max:
        - 15.0
        - 15.0
        - 15.0

# More controller specifications here
# ...