Collaborative Trajectory Prediction via Late Fusion

2026-04-24

AI Summary

This work addresses the uncertainty in trajectory prediction for autonomous driving caused by occlusion, limited sensor range, and perception errors. The authors propose a model-agnostic, asynchronous late-fusion collaborative framework that shifts vehicle-to-vehicle cooperation from the perception stage to the prediction stage. By exchanging low-dimensional prediction outputs instead of high-dimensional features, the approach substantially reduces communication overhead while enabling asynchronous interaction. Evaluated on the OPV2V, V2V4Real, and DeepAccident datasets, the method consistently lowers miss rates and improves trajectory success rates. Notably, on the real-world V2V4Real dataset, it achieves absolute improvements of 1.69% and 1.22% in trajectory success rates for two vehicles, demonstrating its practical deployment potential.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Computer Vision: Vision for Robotics & Autonomous Driving
• Multiagent Systems: Coordination and Collaboration

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Predicting future trajectories of surrounding traffic agents is critical for safe autonomous navigation and collision avoidance. Despite all advances in the trajectory forecasting realm, the prediction models remains vulnerable to uncertainty caused by occlusions, limited sensing range, and perception errors. Collaborative vehicle-to-vehicle (V2V) approaches help reduce this uncertainty by sharing complementary information. Existing collaborative trajectory prediction methods typically fuse feature maps at the perception stage to construct a holistic scene view. Further this holistic representation is decoded into the future trajectories. Such design incurs substantial communication overhead due to the exchange of high-dimensional feature representations and often assumes idealized bandwidth and synchronization, limiting practical deployment. We address these limitations by shifting collaboration from perception to the prediction module and introducing a late-fusion framework for shared forecasts. The framework is model-agnostic and treats collaborating vehicles as independent asynchronous agents. We evaluate the approach on the OPV2V, V2V4Real, and DeepAccident datasets, comparing individual and collaborative forecasting. Across all datasets, late fusion consistently reduces miss rate and improves trajectory success rate ($\mathrm{TSR}_{0.5}$), defined as the fraction of ground-truth agents with final displacement error below 0.5 m. On the real-world V2V4Real dataset, collaborative prediction improves the success rate by $1.69\%$ and $1.22\%$ for both intelligent vehicles, respectively, compared with individual forecasting.

Problem

• collaborative trajectory prediction
• late fusion
• V2V communication
• communication overhead
• autonomous driving

Innovation

• late fusion
• collaborative trajectory prediction
• V2V communication
• model-agnostic framework
• asynchronous agents