Zero-Order Optimization for Gaussian Process-based Model Predictive Control
This work addresses a major bottleneck in learning-based control for real-world applications, offering incremental improvements in computational efficiency.
The paper tackles the challenge of real-time solving for Gaussian process-based model predictive control by employing a tailored Jacobian approximation and parallelizable GP inference, reducing computational complexity from O(n_x^6) to O(n_x^3) per iteration and achieving drastically reduced computation times with feasible solutions.
By enabling constraint-aware online model adaptation, model predictive control using Gaussian process (GP) regression has exhibited impressive performance in real-world applications and received considerable attention in the learning-based control community. Yet, solving the resulting optimal control problem in real-time generally remains a major challenge, due to i) the increased number of augmented states in the optimization problem, as well as ii) computationally expensive evaluations of the posterior mean and covariance and their respective derivatives. To tackle these challenges, we employ i) a tailored Jacobian approximation in a sequential quadratic programming (SQP) approach, and combine it with ii) a parallelizable GP inference and automatic differentiation framework. Reducing the numerical complexity with respect to the state dimension $n_x$ for each SQP iteration from $\mathcal{O}(n_x^6)$ to $\mathcal{O}(n_x^3)$, and accelerating GP evaluations on a graphical processing unit, the proposed algorithm computes suboptimal, yet feasible solutions at drastically reduced computation times and exhibits favorable local convergence properties. Numerical experiments verify the scaling properties and investigate the runtime distribution across different parts of the algorithm.