Task-Priority Kinematic Control for a Mobile Manipulator

Collaborator(s): Atharva Patwe

Code

Report

This project describes the design and implementation of a kinematic control system for a differential-drive robot (Kobuki Turtlebot 2) equipped with a 4 DOF manipulator (uFactory uArm Swift Pro). The system utilizes the Task-Priority Redundancy Resolution Algorithm to manage the robot’s redundant degrees of freedom and to prioritize tasks based on their importance.

The objective of this project is to enable the Kobuki Turtlebot and its manipulator to perform various tasks while ensuring that higher-priority tasks are executed first. The system focuses on managing multiple tasks efficiently in a simulated environment, with potential for real-world applications in autonomous robotics. The implementation was done using the Robot Operating System (ROS) and validated within the Stonefish simulator to test control strategies and task execution.

Methodology

The project followed several key steps in implementation:

• Kinematic Modelling: Developed a kinematic model for both the Kobuki Turtlebot 2 and the uFactory uArm Swift Pro, enabling precise motion control and task execution.
• Task-Priority Redundancy Resolution: Applied the task-priority redundancy resolution algorithm to prioritize tasks effectively and manage the robot’s redundant degrees of freedom. Higher-priority tasks were executed first while still managing lower-priority ones.
• Behavior Trees: Implemented behavior trees to manage task complexity, ensuring smooth task transitions and better system control. Behavior trees systematically handle multiple tasks while maintaining the overall robustness of the system.

• ROS and Stonefish Simulation: Leveraged the ROS framework to implement the control algorithms and used the Stonefish simulator for testing and validation of the system. The simulator provided a robust environment to test various task scenarios before transitioning to potential real-world applications.

Results

The system was rigorously tested in the Stonefish simulator, where the Kobuki Turtlebot 2 and uFactory uArm Swift Pro successfully executed a variety of tasks. The task-priority redundancy resolution algorithm ensured that the most important tasks were always given precedence, while the use of behavior trees effectively managed the complexity of the task transitions.

The video below showcases the system in action, demonstrating the implementation of the kinematic control system using the task-priority redundancy resolution algorithm for the mobile manipulator.

The simulation results, supported by the demonstration video, highlight the effectiveness of the task-priority redundancy resolution algorithm in managing the robot’s control system, offering a flexible and scalable solution for mobile manipulator tasks.