TrajEvo: Trajectory Prediction Heuristics Design via LLM-driven Evolution

TL;DR

TrajEvo introduces an LLM-driven evolutionary framework to automatically design and refine trajectory prediction heuristics, achieving superior accuracy and generalization, especially in out-of-distribution scenarios, compared to traditional methods.

Contribution

The paper presents TrajEvo, a novel framework that uses LLMs and evolutionary algorithms to automatically generate and improve trajectory prediction heuristics, enhancing accuracy and OOD generalization.

Findings

Outperforms existing heuristic methods on multiple datasets.

Surpasses deep learning models in out-of-distribution generalization.

Demonstrates effectiveness of LLM-driven heuristic design.

Abstract

Trajectory prediction is a critical task in modeling human behavior, especially in safety-critical domains such as social robotics and autonomous vehicle navigation. Traditional heuristics based on handcrafted rules often lack accuracy and generalizability. Although deep learning approaches offer improved performance, they typically suffer from high computational cost, limited explainability, and, importantly, poor generalization to out-of-distribution (OOD) scenarios. In this paper, we introduce TrajEvo, a framework that leverages Large Language Models (LLMs) to automatically design trajectory prediction heuristics. TrajEvo employs an evolutionary algorithm to generate and refine prediction heuristics from past trajectory data. We propose two key innovations: Cross-Generation Elite Sampling to encourage population diversity, and a Statistics Feedback Loop that enables the LLM to analyze and improve alternative predictions. Our evaluations demonstrate that TrajEvo outperforms existing heuristic methods across multiple real-world datasets, and notably surpasses both heuristic and deep learning methods in generalizing to an unseen OOD real-world dataset. TrajEvo marks a promising step toward the automated design of fast, explainable, and generalizable trajectory prediction heuristics. We release our source code to facilitate future research at https://github.com/ai4co/trajevo.