Receding horizon control of vectored thrust flight experiment

Abstract

The application of a constrained receding horizon control technique to stabilise an indoor vectored-thrust flight experiment, known as the Caltech ducted fan, is given. The receding horizon control problem is formulated as a constrained optimal control problem and solved in real time with an efficient, computational method that combines nonlinear control theory, B-spline basis functions, and nonlinear programming. Characteristic issues, including non-zero computational times, convergence properties, choice of horizon length and terminal cost are discussed. The study validates the applicability of real-time receding horizon control for constrained systems with fast dynamics.

Real-Time Optimized Rendezvous on Nonholonomic Resource-Constrained Robots

“…Although omitted here for conciseness, it can be shown that this control law is stable and converges to the desired trajectory. This strategy bears resemblance to the third strategy proposed by Milam et al [21] to account for computation delays, with the exception that, instead of blindly applying the optimized control inputs (open-loop), we compute corrected control inputs based on the optimized trajectory using a tracking layer (closed-loop).…”

“…It is an advanced method, widely used in industry, that has the ability to use the available information on the system at hand to control it optimally under a user-defined cost. Although its requirements in terms of computing power are high [21], it has found many successful applications, in particular when the underlying system to control has slow dynamics (i.e., in the order of minutes or seconds). The recent advances in computing power have in part alleviated this issue, but RHC reaches its limits when the underlying system is nonlinear, changing fast and has to run on a simple mobile platform.…”

“…Set II The second set aims to test the degradation of performance in Scenario (a) when the computational time of the controllers is increased by a fixed additional delay of 0, 200 and 400ms. Additionally, we test our approach to mitigate computational delays (Section 2.4, closed-loop) against the third strategy proposed in [21] (open-loop). The performance is measured both through the smoothness and the convergence speed towards the rendezvous point.…”

“…(3) [21]. Hence, it is important that this optimization takes as little time as possible (to guarantee, in practice, the unbiasedness of the estimator ofx i j ).…”

UAV as a Reliable Wingman: A Flight Demonstration

“…The main contribution of this brief is to prove the feasibility of our approach to both nonlinear receding horizon control and formal software modeling and analysis by implementing them on a real flight system operating in an uncertain environment. This flight experiment built on our previous work on applying these tools separately in simulation [5] and on smaller scale experimental testbeds [2], [6], but was the first time they have been joined in a single system. We demonstrated how these two control technologies can work in concert to provide robustness to uncertainty and failures in a dynamic environment and provided a proof-of-concept for how this approach can lead to realizing the goal of human-piloted and autonomous aircraft working together safely.…”