Versatile multicontact planning and control for legged loco-manipulation

Abstract

Loco-manipulation planning skills are pivotal for expanding the utility of robots in everyday environments. These skills can be assessed on the basis of a system’s ability to coordinate complex holistic movements and multiple contact interactions when solving different tasks. However, existing approaches have been merely able to shape such behaviors with hand-crafted state machines, densely engineered rewards, or prerecorded expert demonstrations. Here, we propose a minimally guided framework that automatically discovers whole-body trajectories jointly with contact schedules for solving general loco-manipulation tasks in premodeled environments. The key insight is that multimodal problems of this nature can be formulated and treated within the context of integrated task and motion planning (TAMP). An effective bilevel search strategy was achieved by incorporating domain-specific rules and adequately combining the strengths of different planning techniques: trajectory optimization and informed graph search coupled with sampling-based planning. We showcase emergent behaviors for a quadrupedal mobile manipulator exploiting both prehensile and nonprehensile interactions to perform real-world tasks such as opening/closing heavy dishwashers and traversing spring-loaded doors. These behaviors were also deployed on the real system using a two-layer whole-body tracking controller.

Dynamic Fall Recovery Control for Legged Robots via Reinforcement Learning

“…Although our current approach can generate adaptive control strategies, it is devoid of visual and haptic perception, which are critical for long-term motor planning [63] and dynamic manipulation [64]. To obtain more advanced motor intelligence in unstructured environments, future research needs to integrate visual cues and haptic perception to develop environmentally aware movement.…”

Loco-manipulation with non-impulsive contact-implicit planning in a slithering robot

“…These approaches have found widespread application across various locomotion modalities, such as legged and slithering locomotion, showcasing remarkable efficacy, including rapid contact planning in terrestrial environments [14,8,15]. Notably, point-contact models such as legged robots have been particularly receptive to optimization techniques [87] [1] compared to systems characterized by extensive contact interactions, such as snake robots. Given the intricate dynamics inherent in slithering systems, which encompass sophisticated contact dynamics [22,10,88], there arises a pressing need for enhanced modeling and control methodologies.…”