Selected work

Representative Papers

A Real-Time Generalized Nash Equilibrium Framework for Interaction-Aware Autonomous Driving in Mixed Traffic

Jul 23, 2026

This work addresses the safety and efficiency challenges faced by autonomous vehicles in mixed traffic, where the unpredictable behavior of human drivers complicates interaction. To tackle this issue, the paper formulates driving interactions as a generalized Nash equilibrium problem—a framework introduced here for the first time in a real-time autonomous driving system. By jointly optimizing vehicle strategies while explicitly coupling safety and geometric constraints, the approach enables dynamic coordination between autonomous policies and human behaviors. To solve the resulting non-convex equilibrium problem efficiently, the authors develop a dedicated particle swarm optimization (PSO)-based solver capable of millisecond-level decision-making. Real-world vehicle tests demonstrate that the system converges within 50 milliseconds, generating smooth, human-like trajectories that effectively handle critical interactive scenarios.

0 citationsRead paper

Mapping urban air quality using mobile and fixed low cost sensors: a model comparison

Nov 27, 2025

This study addresses the need for high-resolution urban air quality monitoring by developing low-cost sensor data fusion methodologies. To overcome systematic biases arising from uncalibrated mobile sensors (e.g., vehicular), we propose a “calibration-first, fusion-driven” modeling paradigm: mobile sensor measurements are dynamically calibrated prior to spatiotemporal fusion with reference-grade fixed-station data. Ten statistical and machine learning models are constructed and rigorously evaluated via cross-validation and validation against independent ground-truth station data for both spatial interpolation and concentration forecasting. Results demonstrate that machine learning models achieve the highest predictive accuracy; however, all models exhibit consistent bias when trained on uncalibrated mobile data alone. Integrating calibrated fixed-station data substantially improves estimation accuracy, robustness, and reliability. This work establishes a reproducible technical framework and methodological foundation for high-fidelity urban air quality mapping using heterogeneous, low-cost sensor networks.

0 citationsRead paper

Evaluation of an Uncertainty-Aware Late Fusion Algorithm for Multi-Source Bird's Eye View Detections Under Controlled Noise

Jul 04, 2025

Evaluating multi-sensor bird’s-eye view (BEV) detection fusion performance independently of detector errors remains challenging. To address this, we propose a noise-controllable, systematic evaluation framework and design UniKF—a Kalman filter-based, uncertainty-aware unified post-fusion algorithm. Our framework decouples detector errors from fusion performance via controlled noise injection into BEV feature maps or detections. UniKF explicitly models sensor-specific uncertainties (e.g., calibration and measurement noise) and temporal synchronization offsets, enabling robust cross-sensor fusion under realistic conditions. Experiments across varying noise levels demonstrate that UniKF reduces localization and orientation errors by approximately 3× and size estimation error by 2×, while maintaining near-perfect detection accuracy (99.5%–100% precision and recall). These results significantly outperform state-of-the-art fusion methods, validating both the efficacy of our evaluation framework and the robustness of UniKF under uncertainty.

0 citationsRead paper

A Late Collaborative Perception Framework for 3D Multi-Object and Multi-Source Association and Fusion

Jul 03, 2025

To address the practical bottlenecks in autonomous driving cooperative perception—namely, high communication bandwidth requirements and the need to share model architectures and parameters (compromising privacy)—this paper proposes a lightweight late-fusion framework that relies solely on shared 3D bounding boxes (i.e., class, position, size, and orientation). To our knowledge, this is the first method enabling accurate cross-heterogeneous-system 3D object fusion without accessing agents’ detection model structures or weights. We design a multi-object association and optimization-based fusion algorithm grounded in geometric and semantic consistency, achieving robust matching and error correction at the fusion layer. Experiments demonstrate substantial improvements: position, scale, and orientation estimation errors are reduced to 1/5, 1/7.5, and 1/2 of baseline levels, respectively. Moreover, the method achieves 100% precision and recall in heterogeneous system fusion, while ensuring computational efficiency, strong generalizability, and strict model privacy preservation.

0 citationsRead paper

Recent publications

Latest Papers

A Real-Time Generalized Nash Equilibrium Framework for Interaction-Aware Autonomous Driving in Mixed Traffic

Jul 23, 2026

This work addresses the safety and efficiency challenges faced by autonomous vehicles in mixed traffic, where the unpredictable behavior of human drivers complicates interaction. To tackle this issue, the paper formulates driving interactions as a generalized Nash equilibrium problem—a framework introduced here for the first time in a real-time autonomous driving system. By jointly optimizing vehicle strategies while explicitly coupling safety and geometric constraints, the approach enables dynamic coordination between autonomous policies and human behaviors. To solve the resulting non-convex equilibrium problem efficiently, the authors develop a dedicated particle swarm optimization (PSO)-based solver capable of millisecond-level decision-making. Real-world vehicle tests demonstrate that the system converges within 50 milliseconds, generating smooth, human-like trajectories that effectively handle critical interactive scenarios.

0 citationsRead paper

Mapping urban air quality using mobile and fixed low cost sensors: a model comparison

Nov 27, 2025

This study addresses the need for high-resolution urban air quality monitoring by developing low-cost sensor data fusion methodologies. To overcome systematic biases arising from uncalibrated mobile sensors (e.g., vehicular), we propose a “calibration-first, fusion-driven” modeling paradigm: mobile sensor measurements are dynamically calibrated prior to spatiotemporal fusion with reference-grade fixed-station data. Ten statistical and machine learning models are constructed and rigorously evaluated via cross-validation and validation against independent ground-truth station data for both spatial interpolation and concentration forecasting. Results demonstrate that machine learning models achieve the highest predictive accuracy; however, all models exhibit consistent bias when trained on uncalibrated mobile data alone. Integrating calibrated fixed-station data substantially improves estimation accuracy, robustness, and reliability. This work establishes a reproducible technical framework and methodological foundation for high-fidelity urban air quality mapping using heterogeneous, low-cost sensor networks.

0 citationsRead paper

Evaluation of an Uncertainty-Aware Late Fusion Algorithm for Multi-Source Bird's Eye View Detections Under Controlled Noise

Jul 04, 2025

Evaluating multi-sensor bird’s-eye view (BEV) detection fusion performance independently of detector errors remains challenging. To address this, we propose a noise-controllable, systematic evaluation framework and design UniKF—a Kalman filter-based, uncertainty-aware unified post-fusion algorithm. Our framework decouples detector errors from fusion performance via controlled noise injection into BEV feature maps or detections. UniKF explicitly models sensor-specific uncertainties (e.g., calibration and measurement noise) and temporal synchronization offsets, enabling robust cross-sensor fusion under realistic conditions. Experiments across varying noise levels demonstrate that UniKF reduces localization and orientation errors by approximately 3× and size estimation error by 2×, while maintaining near-perfect detection accuracy (99.5%–100% precision and recall). These results significantly outperform state-of-the-art fusion methods, validating both the efficacy of our evaluation framework and the robustness of UniKF under uncertainty.

0 citationsRead paper

A Late Collaborative Perception Framework for 3D Multi-Object and Multi-Source Association and Fusion

Jul 03, 2025

To address the practical bottlenecks in autonomous driving cooperative perception—namely, high communication bandwidth requirements and the need to share model architectures and parameters (compromising privacy)—this paper proposes a lightweight late-fusion framework that relies solely on shared 3D bounding boxes (i.e., class, position, size, and orientation). To our knowledge, this is the first method enabling accurate cross-heterogeneous-system 3D object fusion without accessing agents’ detection model structures or weights. We design a multi-object association and optimization-based fusion algorithm grounded in geometric and semantic consistency, achieving robust matching and error correction at the fusion layer. Experiments demonstrate substantial improvements: position, scale, and orientation estimation errors are reduced to 1/5, 1/7.5, and 1/2 of baseline levels, respectively. Moreover, the method achieves 100% precision and recall in heterogeneous system fusion, while ensuring computational efficiency, strong generalizability, and strict model privacy preservation.

0 citationsRead paper