Representing Robot Task Plans as Robust Logical-Dynamical Systems

Abstract

It is difficult to create robust, reusable, and reactive behaviors for robots that can be easily extended and combined. Frameworks such as Behavior Trees are flexible but difficult to characterize, especially when designing reactions and recovery behaviors to consistently converge to a desired goal condition. We propose a framework which we call Robust Logical-Dynamical Systems (RLDS), which combines the advantages of task representations like behavior trees with theoretical guarantees on performance. RLDS can also be constructed automatically from simple sequential task plans and will still achieve robust, reactive behavior in dynamic real-world environments. In this work, we describe both our proposed framework and a case study on a simple household manipulation task, with examples for how specific pieces can be implemented to achieve robust behavior. Finally, we show how in the context of these manipulation tasks, a combination of an RLDS with planning can achieve better results under adversarial conditions.

Convergence Analysis of Hybrid Control Systems in the Form of Backward Chained Behavior Trees

“…Then we make some initial assumptions and definitions, followed by the main result. Finally we give some detailed comments on how this result differs from the related work in [10], [11].…”

“…These conditions are added as running conditions for all actions. Thus, [11] is closely related to this work, but this letter goes beyond the work in [11] by similarly identifying conditions that need to hold, but with a higher resolution in time and space. Alternative ways to achieve the same objective can be included, with separate constraints.…”

“…The problem of convergence analysis of BTs has been addressed in a number of recent papers, [4], [5], [10], [11]. In general, all of them analyze a linear chain of actions, each satisfying the preconditions of the following.…”

“…In [11], the authors define a version of BTs called Robust Logical-Dynamical Systems (RLDS) that they use to prove convergence in the presence of disturbances on the task level, such as someone closing a drawer that was opened by the robot. The work is similar in spirit to [10].…”

“…But after placing the object, there is obviously no need to be careful not to drop it, whereas the importance of staying out of the un-safe zone remains. A detailed comparison to the results of [11] can be found in Section IV-D.…”