Locomotion with Online Awareness of Dynamics for Precision Footholds
We present LOAD, a perceptive locomotion framework that combines explicit inertial estimates with implicit dynamics features for narrow-beam traversal.
ProofPaper ↗
Key points
- Unknown payloads alter robot dynamics and make precise contact execution difficult when foothold margins are small.
- The results show that implicit dynamics features accelerate training, while explicit inertial estimation further improves traversal success; combining both enables LOAD to approach the performance of a privileged policy supplied with ground-truth inertial parameters.
- Hardware experiments demonstrate traversal with unilateral payloads of 6 kg on a 10 cm beam and 9 kg on a 15 cm beam, and an evenly distributed 11 kg payload on a 15 cm beam.
- These results support combining physical estimates and history features for precision locomotion under unknown loads.
Sources (1)
- [1]Locomotion with Online Awareness of Dynamics for Precision FootholdsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 13, 08:16 AM
We present LOAD, a perceptive locomotion framework that combines explicit inertial estimates with implicit dynamics features for narrow-beam traversal.
Unknown payloads alter robot dynamics and make precise contact execution difficult when foothold margins are small.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 13, 2026Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent Skills
- Oct 13, 2026GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning
- Oct 13, 2026Simplifying and Improving On-Policy Distillation for Few-Step Diffusion
- Oct 12, 2026[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
- Oct 12, 2026Strengthening our defenses against distillation attacks
- Oct 12, 2026LLM-to-Flow Distillation: Teaching Natural Language Goals to Scientific Generators