GPU Accelerated Convex Approximations for Fast Multi-Agent Trajectory Optimization
This work addresses the challenge of real-time trajectory planning for multiple agents, such as in robotics or autonomous systems, by providing a faster and more efficient solution, though it is incremental as it builds on existing optimization techniques with GPU acceleration.
The paper tackles the problem of fast multi-agent trajectory optimization by developing a computationally efficient optimizer that exploits GPUs to compute trajectories for tens of agents in under a second, showing substantial improvements in computation time and trajectory quality over state-of-the-art methods.
In this paper, we present a computationally efficient trajectory optimizer that can exploit GPUs to jointly compute trajectories of tens of agents in under a second. At the heart of our optimizer is a novel reformulation of the non-convex collision avoidance constraints that reduces the core computation in each iteration to that of solving a large scale, convex, unconstrained Quadratic Program (QP). We also show that the matrix factorization/inverse computation associated with the QP needs to be done only once and can be done offline for a given number of agents. This further simplifies the solution process, effectively reducing it to a problem of evaluating a few matrix-vector products. Moreover, for a large number of agents, this computation can be trivially accelerated on GPUs using existing off-the-shelf libraries. We validate our optimizer's performance on challenging benchmarks and show substantial improvement over state of the art in computation time and trajectory quality.