Multi-mode Trajectory Optimization for Impact-aware Manipulation

Abstract

The transition from free motion to contact is a challenging problem in robotics, in part due to its hybrid nature. Yet, disregarding the effects of impacts at the motion planning level might result in intractable impulsive contact forces. In this paper, we introduce an impact-aware multimodal trajectory optimization (TO) method that comprises both hybrid dynamics and hybrid control in a coherent fashion.A key concept is the incorporation of an explicit contact force transmission model in the TO method. This allows the simultaneous optimization of the contact forces, contact timings, continuous motion trajectories and compliance, while satisfying task constraints. We compare our method against standard compliance control and an impact-agnostic TO method in physical simulations. Further, we experimentally validate the proposed method with a robot manipulator on the task of halting a large-momentum object.

Robot-Safe Impacts with Soft Contacts Based on Learned Deformations

“…Approaches with floating-base legged robots typically try to impose robust control that compensates for non-modeled soft contact properties, for example, with a quadruped in [22] and with a humanoid in [23], [24]. As an alternative to robust control schemes, [25] proposed an impact-aware planning method.…”

Non-prehensile Planar Manipulation via Trajectory Optimization with Complementarity Constraints

“…Recent works [6], [15]- [17] have developed TO methods for planning robot manipulation tasks requiring contact changes. The underlying formulation of these works follows one of two classes: contact-implicit [18] or multi-mode [6], [19]. Here, we focus on the former which expresses the hybrid nature of the contact change as a Mathematical Program with Complementarity Constraints (MPCC).…”

Predicting Impact-Induced Joint Velocity Jumps on Kinematic-Controlled Manipulator

“…Preemptive actions are necessary to avoid violating the hardware-dependent joint velocity bounds. Unlike specifying pre-and post-impact velocity references in joint space [12] or planning impacts carefully [13], it is generally more convenient to formulate quadratic optimization problems in taskspace to meet multiple control objectives and constraints [14]. We preliminarily enabled on-purpose impacts to second order dynamic QP control schemes [1], [15].…”