Gradient Aware - Shrinking Domain based Control Design for Reactive Planning Frameworks used in Autonomous Vehicles

In this paper, we present a novel control law for longitudinal speed control\nof autonomous vehicles. The key contributions of the proposed work include the\ndesign of a control law that reactively integrates the longitudinal surface\ngradient of road into its operation. In contrast to the existing works, we\nfound that integrating the path gradient into the control framework improves\nthe speed tracking efficacy. Since the control law is implemented over a\nshrinking domain scheme, it minimizes the integrated error by recomputing the\ncontrol inputs at every discretized step and consequently provides less\nreaction time. This makes our control law suitable for motion planning\nframeworks that are operating at high frequencies. Furthermore, our work is\nimplemented using a generalized vehicle model and can be easily extended to\nother classes of vehicles. The performance of gradient aware-shrinking domain\nbased controller is implemented and tested on a stock electric vehicle on which\na number of sensors are mounted. Results from the tests show the robustness of\nour control law for speed tracking on a terrain with varying gradient while\nalso considering stringent time constraints imposed by the planning framework.\n