Indirect Multithread Adaptation for Aerospace Robotics Applications

In this paper, a new indirect multithread attracting manifold adaptive controller for a general class of aerospace mechanical/robotic systems is presented. The controller certifies closed-loop stability for any given bounded reference trajectory while adapting many estimates of unknown parameters such that any one particular estimate almost never moves further away from the unknown truth in a distance measure when compared with its current estimate. While maintaining this “no-regret” learning feature for each individual thread, all threads are mixed into a single composite estimate using a judiciously designed weighting scheme that rewards past performance of each individual thread. This provides faster error elimination and superior transient response performance compared with traditional adaptive control implementations that utilize only a single thread. The need for rapid convergence of poorly determined parameters is particularly relevant to on-orbit servicing applications such as capturing non-cooperative space objects and debris. The beneficial features of the proposed multithread learning scheme are demonstrated via numerical simulations of various prototype problems ranging from robotic manipulators to spacecraft control applications.