CusADi: A GPU Parallelization Framework for Symbolic Expressions and Optimal Control
Se Hwan Jeon, Seungwoo Hong, Ho Jae Lee, Charles Khazoom, Sangbae Kim · IEEE Robotics and Automation Letters · 2024
The parallelism afforded by GPUs presents significant advantages in training controllers through reinforcement learning (RL). However, integrating model-based optimization into this process remains challenging due to the complexity of formulating and solving optimization problems across thousands of instances. In this work, we presentCusADi, an extension of thecasadisymbolic framework to support the parallelization of arbitrary closed-form expressions on GPUs withCUDA. We also formulate a closed-form approximation for solving general optimal control problems, enabling large-scale parallelization and evaluation of MPC controllers. Our results show a ten-fold speedup relative to similar MPC implementation on the CPU, and we demonstrate the use ofCusADifor various applications, including parallel simulation, parameter sweeps, and policy training.