Shiyao Zhang

In Object Goal Navigation task, existing methods are typically restricted to static and single-floor environments, ignoring cross-floor topologies and dynamic pedestrian, which limits their real-world deployment. To address these limitations, we propose ZONDA, a zero-shot object navigation with dynamic avoidance framework. In particular, ZONDA integrates three core components: (i) Heuristic multi-floor planning: from height-difference traversable maps, enables stair traversal and cross-floor exploration without a platform-specific learned controller; (ii) Multi-view target verification: cross-checks multi-scale observations with a vision-language model, significantly reducing false positives; and (iii) Dynamic pedestrian avoidance: explicitly tracks and predicts moving pedestrians to generate anticipatory behaviors. Evaluated on a real Direct Drive Tech TITA biped robot and extensive simulations on HM3D and MP3D, ZONDA achieves significantly improved results. Moreover, ZONDA can maintain robust navigation on the dynamic benchmark HM3D-DYNA compared to the existing baseline.

Authors: Shaomin Liang, Xuanhong Liao, and Shiyao Zhang

Paper Link: https://arxiv.org/abs/2607.21025

In future intelligent transportation systems, autonomous cooperative planning (ACP), becomes a promising technique to increase the effectiveness and security of multi-vehicle interactions. However, multiple uncertainties cannot be fully addressed for existing ACP strategies, e.g. perception, planning, and communication uncertainties. To address these, a novel deep reinforcement learning-based autonomous cooperative planning (DRLACP) framework is proposed to tackle various uncertainties on cooperative motion planning schemes. Specifically, the soft actor-critic (SAC) with the implementation of gate recurrent units (GRUs) is adopted to learn the deterministic optimal time-varying actions with imperfect state information occurred by planning, communication, and perception uncertainties. In addition, the real-time actions of autonomous vehicles (AVs) are demonstrated via the Car Learning to Act (CARLA) simulation platform. Evaluation results show that the proposed DRLACP learns and performs cooperative planning effectively, which outperforms other baseline methods under different scenarios with imperfect AV state information.

Authors: Shiyao Zhang, Liwei Deng, Shuyu Zhang, Weijie Yuan, and Hong Zhang

Paper Link: https://ieeexplore.ieee.org/document/11204522/

Cooperative autonomous driving (AD) systems have increasingly become key elements of future intelligent transportation systems owing to the provisioning of dependable, safe, and effective urban mobility operations. In particular, the utilization of motion prediction can contribute to achieving a high-performance cooperative AD planning strategy of the vehicle platoon system. However, realizing accurate spatial-temporal motion prediction is a challenge since most existing work unilaterally considers the spatial or temporal feature in predicting vehicle motion trajectories. To address the problem, we design a novel spatial-temporal Transformer (ST-Transformer) motion prediction model to predict vehicle motion trajectories with highfidelity simulator. In particular, we integrate both the convolutional and transformer-based networks to capture the spatial-temporal feature of vehicle states. Case studies demonstrate the superiority of the proposed model in predicting autonomous vehicle (AV) trajectories over the existing baseline models, which can greatly support AV motion planning tasks.

Authors: Shiyao Zhang, Shuyu Zhang, Song Wang, and Shuangyang Li

Paper Link: https://ieeexplore.ieee.org/document/11174822

Autonomous cooperative planning (ACP) is a promising technique to improve the efficiency and safety of multi-vehicle interactions for future intelligent transportation systems. However, realizing robust ACP is a challenge due to the aggregation of perception, motion, and communication uncertainties. This paper proposes a novel multi-uncertainty aware ACP (MUACP) framework that simultaneously accounts for multiple types of uncertainties via regularized cooperative model predictive control (RC-MPC). The regularizers and constraints for perception, motion, and communication are constructed according to the confidence levels, weather conditions, and outage probabilities, respectively. The effectiveness of the proposed method is evaluated in the Car Learning to Act (CARLA) simulation platform. Results demonstrate that the proposed MUACP efficiently performs cooperative formation in real time and outperforms other benchmark approaches in various scenarios under imperfect knowledge of the environment.

Authors: Shiyao Zhang, He Li, Shengyu Zhang, Shuai Wang, Derrick Wing Kwan Ng, and Chengzhong Xu

Paper Link: https://arxiv.org/abs/2411.00413