This master's thesis focuses on designing, building, and controlling a specific type of cable-driven parallel robot called an "easy-to-install underconstrained cable robot." Here are the key points of the work:
1. Easy Installation:
- Traditional cable robots are bulky and complex to set up. This thesis aims to create a design that simplifies and speeds up the installation process.
- The proposed robot, named Aras-Cam-2, is designed to have all its components (winches, sensors, power supply, etc.) integrated into the mobile platform.
- This allows for easy deployment: simply attach the cables to pre-defined anchor points and the robot is ready to operate.
2. Underconstrained Design:
- Aras-Cam-2 is designed with fewer actuators (cables) than degrees of freedom. This means it has more ways to move than it has motors to control directly.
- This underconstrained design reduces cost and complexity, but introduces control challenges.
- The thesis acknowledges the trade-off between simplicity and controllability, and focuses on developing appropriate control strategies.
3. Trajectory Design:
- Underconstrained robots are prone to unwanted vibrations due to uncontrolled degrees of freedom.
- A key contribution of the thesis is developing a trajectory design methodology to minimize these vibrations.
- The proposed method focuses on "rest-to-rest" trajectories, ensuring the robot starts and ends at rest, thus reducing vibrations.
- The thesis acknowledges the challenges of designing these trajectories for robots with four cables and proposes alternative approaches for future work.
4. Mathematical Modeling and Simulation:
- The thesis develops detailed mathematical models for the robot's kinematics (motion) and dynamics (forces).
- These models are used to simulate the robot's behavior and validate the proposed trajectory design methodology.
- The simulations demonstrate the effectiveness of the rest-to-rest trajectory design in reducing vibrations, particularly for a three-cable version of the robot.
5. Future Work:
- The thesis acknowledges the limitations of the current trajectory design for the four-cable robot and suggests potential solutions for future investigation.
- Two promising avenues are:
- Variational Integrators: This method incorporates system constraints directly into the trajectory design, potentially improving controllability.
- Dynamic Factor Graphs: This optimization-based technique offers a flexible and efficient way to design trajectories while considering various constraints and factors.
