Wei Xu

Data association in multi-object tracking (MOT) can be viewed as an evidence fusion problem, where heterogeneous cues provide incomplete, uncertain, and sometimes conflicting information for candidate associations. Existing tracking-by-detection (TBD) methods typically compress these cues into a scalar association score before assignment, potentially discarding important structural information in the association evidence. This paper proposes an evidence fusion framework for MOT data association based on belief theory. The framework uses belief masses to represent different forms of association evidence, including support, opposition, residual uncertainty, conflict, and unmatched cases. It further introduces a hierarchical reliability modeling scheme that evaluates cue-level reliability from intra-cue ambiguity and inter-cue disagreement, while estimating association-level reliability for candidate associations. Based on this representation, a two-stage fusion mechanism is developed to combine heterogeneous cues and account for competition among association hypotheses. The proposed framework is designed as a plug-in association module for SORT-like trackers and requires no additional training. Experiments across multiple benchmarks, trackers, and detector backbones show consistent improvements across tracking metrics (e.g., HOTA, AssA, and IDF1), demonstrating the effectiveness of the proposed framework under challenging association conditions.

Developing and evaluating visual perception systems for autonomous vehicles requires data across diverse and adverse driving conditions, yet collecting and annotating such real-world data is costly and often impractical. To address this challenge, we propose a modular, scenario-driven framework for generating synthetic datasets tailored to the evaluation of visual perception functions. The framework aligns with the operational boundaries and detection–response requirements of automated driving functions and comprises three stages: (1) configuring use-case-driven scenarios, (2) generating sensor data and ground truth via simulation, and (3) post-processing to ensure dataset usability. Designed to be generic and flexible, the framework is instantiated and demonstrated through its integration with specific platforms and tools, namely Pro-SiVIC and RTMaps. We evaluate the generated dataset from two perspectives, image fidelity and perception performance under synthetic weather conditions, in comparison to real-world conditions. Furthermore, we train multiple perception models under different learning paradigms, including baseline, transfer-learning, and mixed-training strategies, to examine the influence of synthetic data on robustness. Experimental results demonstrate not only the high quality of the generated data but also its effectiveness in evaluating visual perception functions, as well as its benefit to model robustness and generalization.

For a decade, researchers have focused on the development and deployment of road automated mobility. In the development of autonomous driving embedded systems, several stages are required. The first one deals with the perception layers. The second one is dedicated to the risk assessment, the decision and strategy layers and the optimal trajectory planning. The last stage addresses the vehicle control/command. This paper proposes an efficient solution to the second stage and improves a virtual Cooperative Pilot (Co-Pilot) already proposed in 2012. This paper thus introduces a trajectory planning algorithm for automated vehicles (AV), specifically designed for emergency situations and based on the Autonomous Decision-Support Framework (ADSF) of the EU project Trustonomy. This algorithm is an extended version of Elastic Band (EB) with no fixed final position. A set of trajectory nodes is iteratively deduced from obstacles and constraints, thus providing flexibility, fast computation, and physical realism. After introducing the project framework for risk management and the general concept of ADSF, the emergency algorithm is presented and tested under Matlab software. Finally, the Decision-Support framework is implemented under RTMaps software and demonstrated within Pro-SiVIC, a realistic 3D simulation environment. Both the previous virtual Co-Pilot and the new emergency algorithm are combined and used in a near-accident situation and shown in different risky scenarios.