Advanced robot motion planning in task-constrained applications

About the Project

Project description (max 700 words)

The research project focuses on motion planning for advanced robotic systems, with a main attention on two complementary strands:

1. Underactuated Mechanical Systems

The study of underactuated systems represents a significant challenge in robotics, given the limitation of the number of actuators with respect to the degrees of freedom. This project aims to develop motion planning algorithms that optimize the dynamic behaviour of such systems in complex and task-constrained scenarios. Innovative techniques will be explored for:

• The definition of dynamically valid trajectories.
• Robust control to ensure task execution despite under-actuation.
• The integration of geometric and dynamic approaches to improve the feasibility of solutions.

The foreseen applications include mobile robots, flexible manipulators and robotic systems in space or marine contexts.

2. Minimum Torque Motion Planning

A second focus concerns the formulation of motion planning algorithms oriented to minimize the torque required during the execution of constrained tasks. The research uses Model Predictive Control (MPC) techniques, which allow to:

• Anticipate the dynamic behaviour of the robotic system.
• Optimize trajectories while respecting dynamic constraints and actuation capabilities.
• Improve energy efficiency and longevity of mechanical components.

This approach will be implemented in scenarios where precision and efficiency are critical, such as delicate manipulation, robotic surgery and advanced logistics.

Project objectives

• Develop theoretical models and computational tools for the control and planning of underactuated dynamic systems.
• Integrate Model Predictive Control (MPC)-based planning strategies to reduce torque and optimize energy consumption.
• Validate the proposed algorithms through simulations and real-world applications, using mobile robots, manipulators or customized robotic platforms.

Candidate Profile

We are looking for motivated students with skills in:

• Robotics and system dynamics.
• Mathematical optimization and predictive control.
• Programming (MATLAB, Python, C++).

Experience with simulation software such as ROS, Gazebo or Simulink or Coppelia Sim is an advantage.

The project offers the possibility of international collaborations and access to advanced robotics laboratories for experimental tests.