(Georgia Institute of Technology, 1999-07)
Egerstedt, Magnus B.; Hu, Xiaoming; Stotsky, A.
In this paper, the problem of how to integrate different robotic actions, within a behavior-based framework, is investigated. The case we study is the obstacle negotiation problem, and our approach is based on planned on reference trajectories. These trajectories are chosen in such a way that the combination of a robust path-following behavior, designed in a model independent way, and a reactive obstacle-avoidance behavior would move the robot around an obstacle at a predefined safety distance. This is done while keeping the value of a cost functional low, defined for evaluating the performance of the robot. We exemplify and evaluate this approach on a Nomad 200 mobile robot platform.
(Georgia Institute of Technology, 1999-07)
Egerstedt, Magnus B.; Hu, Xiaoming; Stotsky, A.
A solution to the problem of controlling a car-like nonholonomic robot is proposed using a "virtual" vehicle approach, which is shown to be robust with respect to errors and disturbances. The proposed algorithms are model independent, and the stability analysis is done using a dynamical model, in which, for instance, the side slip angles are taken into account.
(Georgia Institute of Technology, 1998-12)
Egerstedt, Magnus B.; Hu, Xiaoming; Stotsky, A.
A solution to the problem of controlling a car-like nonholonomic robot is proposed using a “virtual” vehicle approach, which is shown to be robust with respect to errors and disturbances. The proposed algorithms are model independent, and the stability analysis is done using a dynamical model in which the side slip angles are taken into account.