LLM labeling trajectory
Table of Contents
Use Large Language Model to automatically label trajectory(s) of agent(s) with a set of classes(aggressive driver, calm driver, sunday driver), with respect to the whole environment(map, other agents).
1. possible caveats
- poor performance - it’s possible that LLM just couldn’t do it well
- grounding of structured trajectory data - it’s possible that LLM cannot perceive/understand trajectory in sequence of (x,y) points well.
2. Related keywords
- [trajectory]
- [prediction]
- [LLM,VLM]
- [perception]
- it is possible that it could be done better with chain of thoughts or some pipeline like in GENSIM: GENERATING ROBOTIC SIMULATION TASKS VIA LARGE LANGUAGE MODELS
- flight Trajectory Reconstruction
- Time-series data prediction