This project investigates trajectory generation and tracking for a UR5 manipulator. In the first task, a cubic joint-space trajectory is computed to move the robot from an initial to a goal configuration in 2 seconds with zero boundary velocities. Each joint is interpolated using third-order polynomials, and a proportional controller with feedforward ensures almost perfect tracking.
The second task considers a scenario with obstacles, requiring a feasible motion plan. A Rapidly-exploring Random Tree (RRT) algorithm generates a collision-free path, which is then smoothed and parameterized using a trapezoidal velocity profile. The resulting trajectory achieves continuous motion with 3-second duration and correct boundary conditions.
Simulations confirm that the robot accurately reaches the desired configurations, highlighting the effectiveness of polynomial interpolation in unconstrained environments and motion planning techniques in constrained ones.
More info