Research Topics

My primary research interests lie in the intersection of control theory and machine learning towards applications to connected and autonomous vehicles (CAVs), nonlinear and complex systems. Some past/ongoing projects are summarized below.

Thrust 1. Predictive Energy Efficiency Optimal Control of Connected Electrified Vehicles

Check the related achievements HERE

[1] National Natural Science Foundation of China (NSFC) Project: Integrated Power & Thermal Management of Connected Hybrid Electric Vehicles (61903152) ¥ 300,000

[2] Jilin Provincial DST Project: Intelligent Vehicle Control System within Domain Control Concept (20220301033GX) ¥ 1 million

[3] Golden Award Champion of 6TH IFAC ECOSM Benchmark Competition (Outer performing 55 Teams from 11 Countries)

[4] Second Prize of Science and Technology Progress Award, Chinese Association of Automation (CAA), 2024

[5] X. Gong, J. Wang, B. Ma, L. Lu, Y. Hu*, and H. Chen, “Real-Time Integrated Power and Thermal Management of Connected HEVs Based on Hierarchical Model Predictive Control,” IEEE/ASME Transactions on Mechatronics, vol. 26, no. 3, pp. 1271–1282, 2021. (IF=6.596)

[6] X. Gong, J. Wang, Y. Hu, M. Sun, R. Wang, Y. Yan, L. Guo, and H. Chen, “A Benchmark Study for Real-Time Optimal Control of Connected HEVs: Design, Integration, and Evaluation,” IEEE Transactions on Transportation Electrification, vol. 10, no. 3, pp. 7591–7603, 2023. (IF=7.2)

[7] J. Wang, X. Gong*, P. Wang, Y. Wang, R. Wang, L. Guo, Y. Hu, and H. Chen, “A Stochastic Predictive Adaptive Cruise Control System with Uncertainty-Aware Velocity Prediction and Parameter Self-Learning,” IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 10, pp. 13900–13913, 2024.

[8] H. Yang, Y. Hu, X. Gong*, R. Cao, L. Guo*, and H. Chen, “Energy Management Strategy for Fuel Cell Hybrid Electric Vehicles Considering The Inaccuracy of Predicted Vehicle Speed,” IEEE Transactions on Transportation Electrification, vol. 10, no. 4, pp. 8246–8262, 2024

[9] X. Gong, H. Wang, M. R. Amini, I. Kolmanovsky, J. Sun. “Modeling and Integrated Power and Thermal Management of a Hybrid Electric Vehicle,”IEEE Conference on Control Technology and Applications, Hongkong, 2019.

[10] M. R. Amini, H. Wang, X. Gong, D. Liao-McPherson, I. Kolmanovsky, and J. Sun, “Cabin and Battery Thermal Management of Connected and Automated HEVs for Improved Energy Efficiency Using Hierarchical Model Predictive Control,” IEEE Transactions on Control Systems Technology, vol. 28, no. 5, pp. 1711–1726, 2019.

Funding and Collaborators:

NSFC

FAW-BESTUNE

Future vehicles are expected to be able to exploit increasingly the connected and automated driving environment for efficient and safe driving. Additionally, connectivity and autonomous driving technologies open up new dimensions for control and optimization of vehicle and powertrain systems. While extensive studies have been carried out on fuel economy optimization for electrified vehicles, the implications of the connected and automated vehicles (CAVs) operation on integrated power and thermal management (TM) have not been fully explored. We focus on developing predictive control strategies enabled by advanced traffic modeling and CAV technologies to deliver power and heating (or cooling) for vehicle systems while achieve fuel saving. The detailed focuses of this research includes the following:

Control-oriented prediction for i-PTM and adaption for improved prediction accuracy;
Bench-mark development for i-PTM to evaluate the ceiling of the energy saving;
Real-time climate control for CAVs in congested traffic.

Thrust 2. Data-Physics Hybrid Control Development for Automotive Hierarchical System

Check the related achievements HERE

[1] HUAWEI Project: Data-Driven Vehicle Dynamics Modeling and Control, ¥ 1.4 million

[2] Best Paper Award, IEEE 14TH Data Driven Control and Learning Systems Conference (DDCLS), 2025

[3] Second Prize of the Jilin Provincial Award for Scientific and Technological Progress, 2020

[4] Book: Y. Hu, X. Gong, L. Zhang, J. Gao. Core Control Algorithms for Gasoline Engine Electronic Control Systems. Beijing: Mechanical Industry Press, 2022. (In Chinese)

[5] C. Zhang, Y. Hu*, L. Xiao, X. Gong*, and H. Chen, “Data-Driven Robust Iterative Learning Predictive Control for MIMO Nonaffine Nonlinear Systems with Actuator Constraints,” IEEE Transactions on Industrial Informatics, vol. 20, no. 7, pp. 9850–9860, 2024.

[6] Y. Hu, C. Zhang, B. Wang, J. Zhao, X. Gong*, J. Gao, and H. Chen, “Noise-Tolerant ZNN-Based Data-Driven Iterative Learning Control for Discrete Nonaffine Nonlinear MIMO Repetitive Systems,” IEEE/CAA Journal of Automatica Sinica, vol. 11, no. 2, pp. 344–361, 2024.

[7] J. Zhao, G. Chang, Y. Sun, Y. Hu, H. Chen, J. Yu, H. Chen, and X. Gong*, “Real-Time Nonlinear Model Predictive Control for Optimizing the Ammonia Coverage Ratio of An SCR System Based on TC-QPSO,” Control Engineering Practice, vol. 132, p. 105409, 2023. (IF=5.4)

[8] Z. Li, H. Chen, H. Liu, P. Wang*, and X. Gong*, “Integrated Longitudinal and Lateral Vehicle Stability Control for Extreme Conditions with Safety Dynamic Requirements Analysis,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 10, pp. 19285–19298, 2022

[9] X. Gong, I. Kolmanovsky, E. Garone, K. Zaseck, and H. Chen*, “Constrained Control of Free Piston Engine Generator Based on Implicit Reference Governor,” Science China Information Sciences, vol. 61, pp. 1–17, 2018

[10] F. Xu, H. Chen*, X. Gong, and Q. Mei, “Fast Nonlinear Model Predictive Control on FPGA Using Particle Swarm Optimization,” IEEE Transactions on Industrial Electronics, vol. 63, no. 1, pp. 310–321, 2015.

Funding and Collaborators:

NSFC

HUAWEI

Modern vehicles are evolving into highly complex, hierarchical systems, integrating numerous subsystems with hard physical constraints such as powertrain system, aftertreatment system and vehicle dynamic system. These systems operate across different layers—from low-level actuator control to system-level coordination—each with distinct dynamics and constraints. As safety, energy efficiency, and performance demands escalate, effective multi-layer coordination becomes critical. However, traditional control architectures based on feedforward calibration and PID feedback loops are no longer sufficient to handle the growing number of actuators and multi-objective requirements. While advanced nonlinear optimal control methods offer theoretical promise, their high computational burden makes real-time implementation on resource-limited automotive ECUs infeasible. This disconnect has kept many solutions confined to academic research, with limited industrial adoption. Addressing these challenges calls for computationally efficient, real-time control frameworks that leverage both physics-based models and data-driven insights, tailored to the hierarchical nature of automotive systems—unlocking practical pathways for next-generation vehicle control.

Thrust 3. Intelligent Generation of Safety-Critical Scenarios for Automated Vehicles Testing and Evaluation

Check the related achievements HERE

[1] Jilin Provincial DRC Project: Research on Large-Language-Models for Automotive Applications (2024C003), ¥ 3 millions

[2] SAIC Industrial Foundation Project: Safety Evaluation for AV with Human-Like Behavior Correlation (2105), ¥ 500,000

[3] Special Prize of Technological Invention Award, CAA, 2023

[4] Z. Liu, X. Gong*, N. Li, Y. Lin, T. Qu, Y. Hu, H. Chen, “Adversarial Driving Behavior Generation via Fuzzy Reward Reinforcement Learning Incorporating Human Risk Cognition,” IEEE Transactions on Intelligent Transportation Systems, 2025.

[5] X. Xia, T. Qu, X. Ma, H. Chen, X. Gong*, “From Failures to Fixes: LLM-Driven Scenario Repair for Self-Evolving Autonomous Driving”, ACM-MM, 2025

[6] Z. Liu, H. Gao, Y. Lin, and X. Gong*, “Enhancing Planning for Autonomous Driving via An Iterative Optimization Framework Incorporating Safety-Critical Trajectory Generation,” Remote Sensing, vol. 16, no. 19, p. 3721, 2024.

[7] J. Liu, X. Gong*, T. Wang, Y. Hu, and H. Chen, “A Proxy-Data-Based Hierarchical Adversarial Patch Generation Method,” Computer Vision and Image Understanding, vol. 246, p. 104066, 2024

[8] T. Wang, Y. Hu, Q. Fang, B. He, X. Gong*, and P. Wang*, “DK-Former: A Hybrid Structure of Deep Kernel Gaussian Process Transformer Network for Enhanced Traffic Sign Recognition,” IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 11, pp. 18561–18572, 2024. (IF=7.9)

[9] Z. Liu, H. Gao, H. Ma, S. Cai, Y. Hu, T. Qu, H. Chen, and X. Gong*, “Adversarial Driving Behavior Generation Incorporating Human Risk Cognition for Autonomous Vehicle Evaluation,” in 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 1609–1614, Detroit USA, 2023.

[10] L. Gong, Z. Liu, Y. Hu, T. Qu, H. Chen, and X. Gong*, “Interactive Generation of Dynamically Feasible Vehicle Trajectories Using Dual-VAE,” IFAC-World Conference, vol. 56, no. 2, pp. 2214–2219, Yokohoma, Japan, 2023.

[11] X. Gong, Y. Guo, Y. Feng, J. Sun, and D. Zhao, “Evaluation of The Energy Efficiency in A Mixed Traffic with Automated Vehicles and Human Controlled Vehicles,” in 2018 21st International Conference on Intelligent Transportation Systems (ITSC), pp. 1981–1986, IEEE, 2018.

Funding and Collaborators: SAIC

Comprehensive, reality‑aligned testing now sits at the heart of safe connected and autonomous vehicle (CAV) deployment. We treat the process as a concept of an advanced driving school: each autonomous vehicle is a student that must repeatedly confront—and learn to correct—its own shortcomings before “graduating” to public roads.. For that school to be effective, it must immerse the vehicle in ever‑evolving, lifelike situations rather than replaying a static catalogue of maneuvers. Two gaps currently prevent this ideal. (1) First, most scenario generators overlook genuine human factors—error‑prone habits, aggressive game‑like moves, misjudged risk—so many real safety‑critical events never surface. (2) Second, despite terabytes of historical SIL/HIL data, there is no intelligent curator that can mine past logs, identify the cases most relevant to the vehicle’s present weaknesses, and refashion them into fresh, targeted tests. We make effort to close both gaps: psychology‑grounded behavior models feed an adversarial scenario engine that provokes long‑tail edge cases, while a self‑evolving curator continually harvests and re‑synthesizes historical traces to match current test goals. Together, they form a “school” that grows alongside its “students,” systematically revealing—and helping to eliminate—latent safety risks in next‑generation autonomous vehicles. This thrust was collaborated with Prof. Nan Li from Tongji University and Prof. Xingjun Ma from Fudan University.

Page updated

Google Sites

Report abuse