ALGAMES: A Fast Solver for Constrained Dynamic Games

TL;DR

ALGAMES is a novel solver for constrained dynamic games that efficiently handles multi-actor trajectory optimization, demonstrating real-time autonomous driving capabilities with high robustness and speed.

Contribution

Introduces ALGAMES, a new solver using augmented Lagrangian and quasi-Newton methods for constrained dynamic games, outperforming existing approaches in speed and robustness.

Findings

Solves complex multi-vehicle scenarios three times faster than DDP-based methods.

Achieves real-time MPC performance at over 60 Hz.

Demonstrates robustness through extensive Monte Carlo simulations.

Abstract

Dynamic games are an effective paradigm for dealing with the control of multiple interacting actors. This paper introduces ALGAMES (Augmented Lagrangian GAME-theoretic Solver), a solver that handles trajectory optimization problems with multiple actors and general nonlinear state and input constraints. Its novelty resides in satisfying the first order optimality conditions with a quasi-Newton root-finding algorithm and rigorously enforcing constraints using an augmented Lagrangian formulation. We evaluate our solver in the context of autonomous driving on scenarios with a strong level of interactions between the vehicles. We assess the robustness of the solver using Monte Carlo simulations. It is able to reliably solve complex problems like ramp merging with three vehicles three times faster than a state-of-the-art DDP-based approach. A model predictive control (MPC) implementation of the algorithm demonstrates real-time performance on complex autonomous driving scenarios with an update frequency higher than 60 Hz.