Adaptive system control runs thousands of real-time models on embedded chip
LLA-MPC on Embedded Hardware: Rapid Adaptive Control with Thousands of Parallel Models
Robotics
Summary
Some machines, like race cars, need very fast and accurate control even when conditions keep changing. The authors improved a control method called LLA-MPC so it can quickly try thousands of different ideas about how a system behaves, all at once, using a small computer inside the machine. This helps the system adjust on the fly, like figuring out tire grip on different surfaces while driving. They tested their method on a small race car platform and showed it can follow fast paths better than a fixed model that doesn't adapt.
What this means in practice
- •For embedded system engineers: Implement adaptive control on resource-constrained onboard computers that run thousands of model simulations in parallel for rapid system identification.
- •For autonomous vehicle developers: Improve vehicle performance and safety by enabling online adaptation to changing physical conditions like tire friction and surface type.
Authors
Henry Z. Liao, Maitham F. AL-Sunni, John M. Dolan
Abstract
We present a generalized implementation of Look-Back and Look-Ahead Adaptive Model Predictive Control (LLA-MPC), a learning-free framework for real-time, rapid adaptive system identification and control. The original formulation was demonstrated only in simulation, for autonomous racing with a fixed model structure. Our implementation has modular dynamics and integrators, and we validate it on the F1TENTH platform under constrained computation and noisy state estimation. The system identifies tire parameters online by evaluating thousands of candidate models in real time on an embedded computer. Experiments with low-friction tires across changing surfaces show that LLA-MPC completes high-speed tracking tasks where a fixed nominal model fails. Code, videos, and our related work are available at: https://lla-control.github.io.