sam-llm: interpretable lane change trajectoryprediction via parametric finetuning

2025-09-03

AI Summary

To address the lack of interpretability and physical consistency in lane-change trajectory prediction for autonomous driving, this paper proposes a SAM-LLM hybrid architecture: a large language model (LLM) performs intent reasoning and outputs kinematic parameters—rather than raw coordinates—while an enhanced Sine-Acceleration Model (SAM) generates continuous, physically plausible trajectories. Task-specific parameterized fine-tuning is conducted separately for lane-keeping and lane-changing scenarios. The method achieves a high intent prediction accuracy of 98.73% while compressing trajectory representation volume by 80%, significantly improving interpretability and computational efficiency. It sets a new state-of-the-art in comprehensive performance among comparable approaches. The core innovation lies in the first end-to-end, interpretable co-modeling framework that tightly integrates LLM-based semantic reasoning with physics-grounded kinematic modeling.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: (Large) Language Models

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

This work introduces SAM-LLM, a novel hybrid architecture that bridges the gap between the contextual reasoning of Large Language Models (LLMs) and the physical precision of kinematic lane change models for autonomous driving. The system is designed for interpretable lane change trajectory prediction by finetuning an LLM to output the core physical parameters of a trajectory model instead of raw coordinates. For lane-keeping scenarios, the model predicts discrete coordinates, but for lane change maneuvers, it generates the parameters for an enhanced Sinusoidal Acceleration Model (SAM), including lateral displacement, maneuver duration, initial lateral velocity, and longitudinal velocity change. This parametric approach yields a complete, continuous, and physically plausible trajectory model that is inherently interpretable and computationally efficient, achieving an 80% reduction in output size compared to coordinate-based methods. The SAM-LLM achieves a state-of-the-art overall intention prediction accuracy of 98.73%, demonstrating performance equivalent to traditional LLM predictors while offering significant advantages in explainability and resource efficiency.

Problem

• Predicts interpretable lane change trajectories using parametric finetuning
• Bridges LLM reasoning with kinematic precision for autonomous driving
• Generates physical trajectory parameters instead of raw coordinates

Innovation

• Hybrid LLM-kinematic architecture for autonomous driving
• Parametric fine-tuning outputs physical trajectory parameters
• Sinusoidal Acceleration Model generates interpretable lane changes