HiMAP: History-aware Map-occupancy Prediction with Fallback

2026-02-19

AI Summary

This work addresses the degradation in motion prediction performance and associated safety risks caused by occlusion, missed detections, and ID switches in multi-object tracking. To this end, the authors propose a trajectory prediction framework that operates without explicit identity association: historical detections are transformed into a spatiotemporally invariant occupancy map, from which a dedicated history query module extracts agent-specific past states to enable multimodal future trajectory forecasting. By eliminating reliance on object identities, the method supports streaming inference and serves as a robust fallback during tracking failures. Evaluated on Argoverse 2, it matches the performance of tracking-based approaches under standard conditions and, in a tracking-free setting, improves Final Displacement Error (FDE) by 11% and Average Displacement Error (ADE) by 12% over QCNet, while reducing Miss Rate (MR) by 4%.

Technology Category

• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

Accurate motion forecasting is critical for autonomous driving, yet most predictors rely on multi-object tracking (MOT) with identity association, assuming that objects are correctly and continuously tracked. When tracking fails due to, e.g., occlusion, identity switches, or missed detections, prediction quality degrades and safety risks increase. We present \textbf{HiMAP}, a tracking-free, trajectory prediction framework that remains reliable under MOT failures. HiMAP converts past detections into spatiotemporally invariant historical occupancy maps and introduces a historical query module that conditions on the current agent state to iteratively retrieve agent-specific history from unlabeled occupancy representations. The retrieved history is summarized by a temporal map embedding and, together with the final query and map context, drives a DETR-style decoder to produce multi-modal future trajectories. This design lifts identity reliance, supports streaming inference via reusable encodings, and serves as a robust fallback when tracking is unavailable. On Argoverse~2, HiMAP achieves performance comparable to tracking-based methods while operating without IDs, and it substantially outperforms strong baselines in the no-tracking setting, yielding relative gains of 11\% in FDE, 12\% in ADE, and a 4\% reduction in MR over a fine-tuned QCNet. Beyond aggregate metrics, HiMAP delivers stable forecasts for all agents simultaneously without waiting for tracking to recover, highlighting its practical value for safety-critical autonomy. The code is available under: https://github.com/XuYiMing83/HiMAP.

Problem

• motion forecasting
• multi-object tracking failure
• trajectory prediction
• autonomous driving
• tracking-free prediction

Innovation

• tracking-free prediction
• historical occupancy maps
• history-aware querying
• DETR-style decoder
• robust motion forecasting