Trajectory Optimization for a Large Number of Autonomous Vehicles

Overview

In this project, I designed an mean field game-based scheme specially for the trajectory optimization for a large number of agents, which is challenging for traditional methods due to complex interactions between agents. Our scheme jointly consider the energy consumption and obstacle avoidance. Moreover, it is applicable for different scenarios, such as autonomous vehicles, UAVs, or intelligent robots.

Experiment

• Robotics swarm experiment is conducted with Robotarium at Georgia Tech
• Numerical simulations: Matlab

Publication

H. Gao et al., “Energy-Efficient Velocity Control for Massive Numbers of UAVs: A Mean Field Game Approach,” in IEEE Transactions on Vehicular Technology, vol. 71, no. 6, pp. 6266-6278, June 2022, doi: 10.1109/TVT.2022.3158896.

Demo Video