Speech-to-Trajectory: Learning Human-Like Verbal Guidance for Robot Motion (2504.05084v1)

Abstract: Full integration of robots into real-life applications necessitates their ability to interpret and execute natural language directives from untrained users. Given the inherent variability in human language, equivalent directives may be phrased differently, yet require consistent robot behavior. While LLMs have advanced language understanding, they often falter in handling user phrasing variability, rely on predefined commands, and exhibit unpredictable outputs. This letter introduces the Directive LLM (DLM), a novel speech-to-trajectory framework that directly maps verbal commands to executable motion trajectories, bypassing predefined phrases. DLM utilizes Behavior Cloning (BC) on simulated demonstrations of human-guided robot motion. To enhance generalization, GPT-based semantic augmentation generates diverse paraphrases of training commands, labeled with the same motion trajectory. DLM further incorporates a diffusion policy-based trajectory generation for adaptive motion refinement and stochastic sampling. In contrast to LLM-based methods, DLM ensures consistent, predictable motion without extensive prompt engineering, facilitating real-time robotic guidance. As DLM learns from trajectory data, it is embodiment-agnostic, enabling deployment across diverse robotic platforms. Experimental results demonstrate DLM's improved command generalization, reduced dependence on structured phrasing, and achievement of human-like motion.