How to achieve various gait patterns from single nominal
Miomir Vukobratovic, Dejan Andric, Branislav Borovac
Abstract
In this paper is presented an approach to achieving on-line modification of nominal biped gait without recomputing entire dynamics when steady motion is performed. Straight, dynamically balanced walk was used as a nominal gait, and applied modifications were speed-up and slow-down walk and turning left and right. It is shown that the disturbances caused by these modifications jeopardize dynamic stability, but they can be simply compensated to enable walk continuation.
Create a lesson
Related papers
GeoAAC: Geometry-Based Adaptive Action Chunking from Denoising Trajectories in VLA Policies
Xin Chen, Sen Chen, Yujuan Ding et al.
Agile-WAM: An Agile Tactile World Action Model for Contact-Rich Robot Control
Hanchu Zhou, Brendan Lynch, Raman Goyal et al.
OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher
Damiano Da Col, Maximilian Igl, Peter Karkus et al.
MILER: Semantic Mid-Level Representation for Sim-to-Real Reinforcement Learning in Unstructured Autonomous Driving
Thomas Steinecker, Denis Trescher, Alexander Bienemann et al.
Underwater Visual Target Tracking with Target-Specific Depth Estimation and Adaptive Model-Fusion Predictive Control
Yuheng Zhou, Haiyang Cheng, Yanqi Feng et al.
MoWAM: Explicit Future Motion Prediction for Efficient World Action Models
Jiayu Wang, Bin Zhu, Yue Yu et al.