PulseAugur
EN
LIVE 09:16:55

Humanoid robots gain parkour skills with new perception and training frameworks

By PulseAugur Editorial · [2 sources] ·

Researchers have developed two new frameworks for enabling humanoid robots to navigate complex, unstructured terrains at high speeds. The first, "Hiking in the Wild," uses a perceptive parkour framework that maps depth and proprioception directly to joint actions, achieving speeds up to 2.5 m/s on a Unitree G1 robot. The second, "TTT-Parkour," employs rapid test-time training in simulation to fine-tune policies for specific, unseen environments, allowing robots to master challenging obstacles within approximately 10 minutes. AI

IMPACT These frameworks could significantly advance the capabilities of humanoid robots in real-world applications, enabling more agile and adaptive movement in unpredictable environments.

RANK_REASON Two research papers detailing new frameworks for humanoid robot locomotion.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Humanoid robots gain parkour skills with new perception and training frameworks

How we ranked this

Signal score
22 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
Two research papers detailing new frameworks for humanoid robot locomotion.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [2]

  1. arXiv cs.AI TIER_1 English(EN) · Shaoting Zhu, Ziwen Zhuang, Mengjie Zhao, Kun-Ying Lee, Hang Zhao ·

    Hiking in the Wild: A Scalable Perceptive Parkour Framework for Humanoids

    arXiv:2601.07718v2 Announce Type: replace-cross Abstract: Achieving robust humanoid locomotion in complex, unstructured environments requires transitioning from reactive proprioception to proactive perception. However, integrating exteroception remains a significant challenge: ex…

  2. arXiv cs.AI TIER_1 English(EN) · Shaoting Zhu, Baijun Ye, Jiaxuan Wang, Jiakang Chen, Ziwen Zhuang, Linzhan Mou, Runhan Huang, Hang Zhao ·

    TTT-Parkour: Rapid Test-Time Training for Perceptive Robot Parkour

    arXiv:2602.02331v2 Announce Type: replace-cross Abstract: Achieving highly dynamic parkour with humanoid robots on complex terrains unseen during the pre-training stage remains a challenge in robotics. Although general locomotion policies demonstrate capabilities across broad ter…

RELATED ENTITIES

RELATED TOPICS