A Thesis on Loco-Manipulation with Non-impulsive Contact-Implicit Planning in a Slithering Robot
Kruthika Gangaraju
TL;DR
This thesis addresses loco-manipulation by a snake-like COBRA robot using non-impulsive contact-implicit path planning to enable object manipulation through coordinated locomotion. It combines a high-fidelity Simscape-based model with open-loop central pattern generator gaits to generate and validate manipulation strategies on flat surfaces and ramps, including lifting and placing boxes and moving them to targets. Key contributions include a comprehensive contact modelling framework, open-loop gait design with multiple rolling and sidewinding modes, and successful hardware replication of simulated results, laying the groundwork for a closed-loop, contact-rich optimization approach. The work advances practical loco-manipulation in unstructured environments and informs future development of perception-informed, feedback-controlled manipulation for mobile robots.
Abstract
Object manipulation has been extensively studied in the context of fixed base and mobile manipulators. However, the overactuated locomotion modality employed by snake robots allows for a unique blend of object manipulation through locomotion, referred to as loco-manipulation. The following work presents an optimization approach to solving the loco-manipulation problem based on non-impulsive implicit contact path planning for our snake robot COBRA. This thesis presents the mathematical framework and show high-fidelity simulation results and experiments to demonstrate the effectiveness of our approach.
