Safe and Efficient Adaptive-Predictive Control of Constrained Nonlinear Systems

Опубликовано 7 ноября 2016, 20:58

As autonomous systems are deployed in increasingly complex and uncertain environments, safe and accurate feedback control techniques are required to ensure reliable operation. Accurate trajectory tracking is essential to complete a variety of tasks, but this may be difficult if the system's dynamics change online, e.g., due to environmental effects or hardware degradation. As a result, uncertainty mitigation techniques are also necessary to ensure safety and accuracy. However, this is particularly difficult for small, agile systems, such as micro aerial vehicles (MAVs), which have severely limited computation due to size, weight, and power restrictions but require high-rate feedback control to maintain stability. In this talk, I will present a computationally-efficient, Nonlinear Model Predictive Control strategy that enables accurate and reliable operation in the presence of unmodeled system dynamics while ensuring safety via constraint satisfaction. This technique uses past experiences to inform an online-updated estimate of the system dynamics model and the choice of controller for a given scenario, thereby reducing online computation and enhancing control performance. We demonstrate the safety and accuracy of this high-rate, constrained, adaptive controller via hardware-in-the-loop simulations and experimental studies with a small quadrotor MAV.

See more on this video at microsoft.com/en-us/research/v...