Advancements in Legged Robot Locomotion and Manipulation

The field of legged robotics is witnessing significant advancements in locomotion and manipulation capabilities. Researchers are exploring innovative approaches to improve the agility, stability, and adaptability of legged robots, enabling them to navigate complex terrains and perform diverse tasks. A key direction is the development of control frameworks that can generalize across different robot morphologies, allowing for more versatile and efficient locomotion. Another area of focus is the integration of reinforcement learning and model predictive control to optimize locomotion and manipulation tasks. Noteworthy papers in this area include: RAKOMO, which proposes a path optimization technique for legged manipulators, and Bipedalism for Quadrupedal Robots, which introduces a risk-adaptive reinforcement learning framework for quadrupedal robots walking on their hind legs. These advancements have the potential to significantly enhance the capabilities of legged robots and expand their applications in various fields.

Sources

RAKOMO: Reachability-Aware K-Order Markov Path Optimization for Quadrupedal Loco-Manipulation

Bipedalism for Quadrupedal Robots: Versatile Loco-Manipulation through Risk-Adaptive Reinforcement Learning

A Nonlinear MPC Framework for Loco-Manipulation of Quadrupedal Robots with Non-Negligible Manipulator Dynamics

FLORES: A Reconfigured Wheel-Legged Robot for Enhanced Steering and Adaptability

UniLegs: Universal Multi-Legged Robot Control through Morphology-Agnostic Policy Distillation

In-between Motion Generation Based Multi-Style Quadruped Robot Locomotion

Quadratic Programming-Based Posture Manipulation and Thrust-vectoring for Agile Dynamic Walking on Narrow Pathways

Built with on top of