EdgeVTP: Exploration of Latency-efficient Trajectory Prediction for Edge-based Embedded Vision Applications

2026-04-17

AI Summary

This work addresses the stringent low-latency and determinism requirements of vehicle trajectory prediction on roadside edge devices in highway scenarios by proposing an efficient, embedded-deployment-oriented method. The approach integrates an interaction-aware graph neural network, a lightweight Transformer backbone, and a single-pass curve decoder, innovatively replacing conventional autoregressive waypoint prediction with anchor-based curve parameterization. Interaction complexity is explicitly constrained through a local graph with a hard upper bound on neighboring agents. Experiments across three highway datasets and two Jetson platforms demonstrate that the method achieves state-of-the-art accuracy on two datasets, delivers competitive performance on all other metrics, and attains the lowest measured end-to-end latency among comparable approaches.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Vehicle trajectory prediction is central to highway perception, but deployment on roadside edge devices necessitates bounded, deterministic end-to-end latency. We present EdgeVTP, an embedded-first trajectory predictor that combines interaction-aware graph modeling with a lightweight transformer backbone and a one-shot curve decoder. By predicting future motion as compact curve parameters (anchored at the last observed position) rather than horizon-scaled autoregressive waypoints, EdgeVTP reduces decoding overhead while producing smooth trajectories. To keep runtime predictable in crowded scenes, we explicitly bound interaction complexity via a locality graph with a hard neighbor cap. Across three highway benchmarks and two Jetson-class platforms, EdgeVTP achieves the lowest measured end-to-end latency under a protocol that includes graph construction and post-processing, while attaining state-of-the-art (SotA) prediction accuracy on two of the three datasets and competitive error on other benchmarks. Our code is available at https://github.com/SeungjinStevenKim/EdgeVTP.

Problem

• trajectory prediction
• edge computing
• latency efficiency
• embedded vision
• deterministic latency

Innovation

• trajectory prediction
• edge computing
• lightweight transformer
• curve parameterization
• interaction-aware graph