Multi-Agent Dynamic Simulator for Lunar Environment
Francesco Borrelli, Professor
Mechanical Engineering
Applications for Fall 2024 are closed for this project.
In order to mitigate risk during deployment, control algorithms for robots on extraterrestrial missions (such as the NASA JPL Artemis rover) must first be evaluated in high-fidelity dynamic simulators. These simulators must accurately model the robot dynamics within a complex, changing environment. The goal of this project is for students to use the ROS 2 framework (or similar) to construct such a simulator which can efficiently and accurately model the dynamics of multiple agents when moving across complex topological features and in changing temperatures. Once designed, the simulator will be used for evaluating next-stage autonomy systems for multi-agent remote missions.
Role: Students will construct a multi-agent dynamic simulator for robots on extraterrestrial missions, using ROS 2 or a similar framework. The simulator must be able to model the dynamics of multiple agents under a variety of control schemes. After the simulator is designed, students will utilize it to develop and test next-stage autonomy control architectures.
Throughout the project, students will grow their skills in sensor fusion, dynamic modeling, simulator architecture design, autonomy systems, and control design.
Qualifications: Coding experience in Python, Matlab or C++ is required. Some familiarity with ROS or ROS 2 would be helpful.
Hours: 9-11 hrs
Mathematical and Physical Sciences Engineering, Design & Technologies