CTBC: Contact-Triggered Blind Climbing for Wheeled Bipedal Robots with Instruction Learning and Reinforcement Learning

2025-09-02

AI Summary

Wheel-legged robots exhibit limited obstacle-crossing capability on unstructured terrain—such as staircases—particularly when obstacles significantly exceed wheel diameter. To address this, we propose the Contact-Triggered Blind Climbing (CTBC) framework, which employs ground-wheel contact signals as an immediate, vision-free trigger to initiate rapid leg lifting. CTBC integrates strongly guided feedforward trajectories with proprioceptive feedback, synergistically combining imitation learning and reinforcement learning to yield a highly responsive, low-latency control policy. Experimentally validated on the Tron1 robot, CTBC enables stable stair climbing up to 2.3× the wheel radius—surpassing conventional physical limits of wheeled platforms—and markedly enhances autonomous navigation and adaptability in complex environments. The core innovation lies in the first use of contact feedback as the initiation mechanism for blind climbing, achieving tight perception–action coupling for agile, real-time obstacle negotiation.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Computer Vision: Low Level & Physics-based Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: The sharing economy

Abstract

In recent years, wheeled bipedal robots have gained increasing attention due to their advantages in mobility, such as high-speed locomotion on flat terrain. However, their performance on complex environments (e.g., staircases) remains inferior to that of traditional legged robots. To overcome this limitation, we propose a general contact-triggered blind climbing (CTBC) framework for wheeled bipedal robots. Upon detecting wheel-obstacle contact, the robot triggers a leg-lifting motion to overcome the obstacle. By leveraging a strongly-guided feedforward trajectory, our method enables the robot to rapidly acquire agile leg-lifting skills, significantly enhancing its capability to traverse unstructured terrains. The approach has been experimentally validated and successfully deployed on LimX Dynamics' wheeled bipedal robot, Tron1. Real-world tests demonstrate that Tron1 can reliably climb obstacles well beyond its wheel radius using only proprioceptive feedback.

Problem

• Enabling wheeled bipedal robots to climb obstacles
• Overcoming performance limitations on complex terrains
• Achieving reliable climbing using proprioceptive feedback only

Innovation

• Contact-triggered blind climbing framework
• Leg-lifting motion triggered by wheel-obstacle contact
• Strongly-guided feedforward trajectory for agile skills