Huihua Gao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Huihua Gao has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Ting Qu
National Key Laboratory of Automotive Chassis Integration and Bionics, Jilin University, Changchun, China
3 shared publications - Hong Chen
Tongji University
2 shared publications - Yunfeng Hu
Jilin University
2 shared publications - Bingzhao Gao
Tongji University
1 shared publication - Fang Xu
Jilin University
1 shared publication - Ping Wang
Jilin University
1 shared publication
Latest Publications
- 28 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. PP (02780046) , 1-13
AdvMPC: A Controllable and Efficient Adversarial MPC Framework for Safety Validation in Autonomous Driving
Generating controllable, efficient, and physically feasible adversarial driving behaviors under dynamic interactions remains a fundamental challenge in the safety validation of autonomous driving systems. Leveraging the advantages of model predictive control (MPC) in receding-horizon optimization and constraint handling, this article proposes adversarial model predictive control (AdvMPC), a controllable and efficient adversarial MPC framework for safety validation in autonomous driving. The proposed framework enables background vehicles to generate continuous, executable, and highly adversarial behaviors under realism constraints. It consists of two core components: an adversarial risk modeling and evaluation module, which quantitatively characterizes interaction risk and enforces scenario rationality; and an optimization formulation and solution module, which enhances controllability, interpretability, and computational efficiency through a stage-guided problem construction strategy and an event-triggered warm-start mechanism. Simulation results demonstrate that AdvMPC can reliably generate challenging safety-critical scenarios across diverse dynamic interactions, while ensuring vehicle dynamics feasibility. Furthermore, the framework has been successfully deployed on a hardware-in-the-loop platform for autonomous driving testing.
- 28 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. PP (02780046) , 1-13
AdvMPC: A Controllable and Efficient Adversarial MPC Framework for Safety Validation in Autonomous Driving
Generating controllable, efficient, and physically feasible adversarial driving behaviors under dynamic interactions remains a fundamental challenge in the safety validation of autonomous driving systems. Leveraging the advantages of model predictive control (MPC) in receding-horizon optimization and constraint handling, this article proposes adversarial model predictive control (AdvMPC), a controllable and efficient adversarial MPC framework for safety validation in autonomous driving. The proposed framework enables background vehicles to generate continuous, executable, and highly adversarial behaviors under realism constraints. It consists of two core components: an adversarial risk modeling and evaluation module, which quantitatively characterizes interaction risk and enforces scenario rationality; and an optimization formulation and solution module, which enhances controllability, interpretability, and computational efficiency through a stage-guided problem construction strategy and an event-triggered warm-start mechanism. Simulation results demonstrate that AdvMPC can reliably generate challenging safety-critical scenarios across diverse dynamic interactions, while ensuring vehicle dynamics feasibility. Furthermore, the framework has been successfully deployed on a hardware-in-the-loop platform for autonomous driving testing.
- Open Access
Driving Intention Recognition of Surrounding Vehicles Based on a Time-Sequenced Weights Hidden Markov Model for Autonomous Driving
Accurate perception, especially situational awareness, is central to the evolution of autonomous driving. This necessitates understanding both the traffic conditions and driving intentions of surrounding vehicles. Given the unobservable nature of driving intentions, the hidden Markov model (HMM) has emerged as a popular tool for intention recognition, owing to its ability to relate observable and hidden variables. However, HMM does not account for the inconsistencies present in time series data, which are crucial for intention recognition. Specifically, HMM overlooks the fact that recent observations offer more reliable insights into a vehicle’s driving intention. To address the aforementioned limitations, we introduce a time-sequenced weights hidden Markov model (TSWHMM). This model amplifies the significance of recent observations in recognition by integrating a discount factor during the observation sequence probability computation, making it more aligned with practical requirements. Regarding the model’s input, in addition to easily accessible states of a target vehicle, such as lateral speed and heading angle, we also introduced lane hazard factors that reflect collision risks to capture the traffic environment information surrounding the vehicle. Experiments on the HighD dataset show that TSWHMM achieves recognition accuracies of 94.9% and 93.4% for left and right lane changes, surpassing both HMM and recurrent neural networks (RNN). Moreover, TSWHMM recognizes lane-changing intentions earlier than its counterparts. In tests involving more complex roundabout scenarios, TSWHMM achieves an accuracy of 87.3% and can recognize vehicles’ intentions to exit the roundabout 2.09 s in advance.
Open AccessDriving Intention Recognition of Surrounding Vehicles Based on a Time-Sequenced Weights Hidden Markov Model for Autonomous Driving
Accurate perception, especially situational awareness, is central to the evolution of autonomous driving. This necessitates understanding both the traffic conditions and driving intentions of surrounding vehicles. Given the unobservable nature of driving intentions, the hidden Markov model (HMM) has emerged as a popular tool for intention recognition, owing to its ability to relate observable and hidden variables. However, HMM does not account for the inconsistencies present in time series data, which are crucial for intention recognition. Specifically, HMM overlooks the fact that recent observations offer more reliable insights into a vehicle’s driving intention. To address the aforementioned limitations, we introduce a time-sequenced weights hidden Markov model (TSWHMM). This model amplifies the significance of recent observations in recognition by integrating a discount factor during the observation sequence probability computation, making it more aligned with practical requirements. Regarding the model’s input, in addition to easily accessible states of a target vehicle, such as lateral speed and heading angle, we also introduced lane hazard factors that reflect collision risks to capture the traffic environment information surrounding the vehicle. Experiments on the HighD dataset show that TSWHMM achieves recognition accuracies of 94.9% and 93.4% for left and right lane changes, surpassing both HMM and recurrent neural networks (RNN). Moreover, TSWHMM recognizes lane-changing intentions earlier than its counterparts. In tests involving more complex roundabout scenarios, TSWHMM achieves an accuracy of 87.3% and can recognize vehicles’ intentions to exit the roundabout 2.09 s in advance.