Strategizing at Speed: A Learned Model Predictive Game for Multi-Agent Drone Racing
For autonomous drone racing, this work provides a method that balances strategic depth and computational speed, enabling better performance in high-speed multi-agent scenarios.
The paper investigates the trade-off between interaction-aware planning (MPG) and fast but non-interactive planning (MPC) in multi-agent drone racing, finding that MPG outperforms at moderate speeds but suffers from latency at high speeds. The proposed Learned Model Predictive Game (LMPG) reduces latency and outperforms both baselines in head-to-head races.
Autonomous drone racing pushes the boundaries of high-speed motion planning and multi-agent strategic decision-making. Success in this domain requires drones not only to navigate at their limits but also to anticipate and counteract competitors' actions. In this paper, we study a fundamental question that arises in this domain: how deeply should an agent strategize before taking an action? To this end, we compare two planning paradigms: the Model Predictive Game (MPG), which finds interaction-aware strategies at the expense of longer computation times, and contouring Model Predictive Control (MPC), which computes strategies rapidly but does not reason about interactions. We perform extensive experiments to study this trade-off, revealing that MPG outperforms MPC at moderate velocities but loses its advantage at higher speeds due to latency. To address this shortcoming, we propose a Learned Model Predictive Game (LMPG) approach that amortizes model predictive gameplay to reduce latency. In both simulation and hardware experiments, we benchmark our approach against MPG and MPC in head-to-head races, finding that LMPG outperforms both baselines.