Articulated tracked robots improve traversal with language guided control
ASTRIL-MPC: Autonomous Traversal Framework of Articulated Tracked Robots with Language-Guided Neural-Kinematic MPC
RoboticsArtificial Intelligence
Summary
Robots that use tracks and movable arms need to navigate tricky places like stairwells during search and rescue missions. The authors created a new system called ASTRIL-MPC that uses learned models and smart planning to move these robots safely and smoothly. It also uses language models to adjust how the robot moves, making it better at avoiding collisions and adapting to different environments. Their experiments show that this method works much better than other control approaches.
What this means in practice
- •For robot operators: Improve robot autonomy and safety in urban search and rescue operations facing complex terrain like stairwells and cluttered interiors.
- •For warehouse automation teams: Enhance navigation control for robots moving in cluttered indoor environments with mixed surfaces and obstacles.
Authors
Zhenfeng Gan, Yanbo Chen, Lirong Che, Junbo Tan, Xueqian Wang
Abstract
In urban search and rescue, articulated tracked robots (ATRs) must traverse structured but contact-rich environments such as stairwells and cluttered building interiors. Reliable autonomy remains challenging because robot-terrain interaction (RTI) is hybrid and discontinuous, and effective flipper-track coordination is difficult to model analytically. We present ASTRIL-MPC, a language-guided neural kinematics model predictive control (MPC) framework for autonomous traversal. A learned kinematics model predicts short-horizon task-state increments from a height sequence and recent trajectories; NMPC plans with multi-objective costs and strict feasibility constraints; and a large language model (LLM) proposes bounded updates to selected weights and bounds through a safety-checked interface with range clipping, rate limiting, and consistency checks. The compiled predictor enables a full control cycle within 100 ms. Across three traversal tasks and a multi-height generalization setting, ASTRIL-MPC improves an aggregate traversal-quality score by up to 71% over a non-adaptive NMPC and by 67% over a PPO baseline, while eliminating measurable collision impacts during descent. These results indicate that combining learned kinematics, optimization-based planning, and language-guided retuning yields data-efficient and robust autonomy for articulated tracked robots.