Liang Yan

Liang Yan

Auto-generated profile, based on publicly available information

Liang Yan

Institute of Intelligent Vehicle, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 July 2024 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
LYLiang Yan
Liang Yan
Xiaodong WuXiaodong Wu
Xiaodong Wu
Chongfeng WeiChongfeng Wei
Chongfeng Wei
SZSheng Zhao
Sheng Zhao
Although artificial intelligence has made tremendous progress recently, there remain various technical issues and ethical problems before autonomous vehicles can be available to the general public. In the interim, collaboration between automated systems and human drivers becomes a promising solution, where the merits of machine intelligence and human intelligence are blended in a complementary way. To this end, the paper proposes a reference-free human-vehicle shared control framework based on reinforcement learning. Firstly, a personalized human-like driver agent is derived from highway driving data by means of generative adversarial imitation learning integrated with Gaussian mixture model. The driver model is responsible for real-time interaction with the reinforcement learning agent to relieve the burden of human operators in the course of training. Then, a copilot agent learns the policies to cooperatively control the vehicle steering based on three distinct implementations for the search of the best coordination strategy. Heuristic reward functions are designed to guide the agent policy optimization for multi-objective equilibrium between driver synchronization against intervention. To verify the control performance of the proposed shared driving system, simulation experiments with driver models and human-in-the-loop tests with real-life participants are conducted in the end of this paper. The results demonstrate that the shared steering control method can effectively follow human intentions, facilitate driving goals, improve road safety and reduce driver’s workload simultaneously in the challenging dynamic obstacle avoidance scenarios.

10 November 2023 in Clinical Microbiology and Infection
Verified
Journal Article
Open access
HHHaifeng Hu
Haifeng Hu
JZJiayi Zhan
Jiayi Zhan
WCWenjing Chen
Wenjing Chen
YYYali Yang
Yali Yang
HJHong Jiang
Hong Jiang
JLJianqi Lian
Jianqi Lian
To develop and validate a simple and effective death risk stratification scale for hemorrhagic fever with renal syndrome (HFRS). In this ambispective cohort study, we investigated the epidemiological and clinical data of 2245 patients with HFRS (1873 enrolled retrospectively and constituting the training cohort, 372 prospectively recruited as the validation cohort) from September 2008 to December 2021, and identified independent risk factors for 30-day death of HFRS. Using logistic regression analysis, a nomogram prediction model was established and was further simplified into a novel scoring scale. Calibration plot, receiver operating characteristic curve, net reclassification index, integrated discrimination index, and decision curve analysis were used to assess the calibration, discrimination, precision, and clinical utility in both training and validation cohorts. Of 2245 patients with HFRS, 132 (5.9%) died during hospitalization. The nomogram prediction model and scoring scale were developed using six predictors: comorbid hypertension, hypotensive shock, hypoxemia, neutrophils, aspartate aminotransferase, and activated partial thromboplastin time. Both the scale and nomogram were well calibrated (near-diagonal calibration curves) and demonstrated significant predictive values (areas under receiver operating characteristic curves >0.9, sensitivity and specificity >90% in the training cohort and >84% in the validation cohort). The simplified scoring scale demonstrated equivalent discriminative ability to the nomogram, with net reclassification index and integrated discrimination index of 0.022 and 0.007 in the training cohort, 0.126 and 0.022 in the validation cohort. Decision curve analysis graphically represented significant clinical utility and comparable net benefits of the nomogram and scoring scale across a range of threshold probabilities. This evidence-based, factor-weighted, accurate score could help clinicians swiftly stratify HFRS mortality risk and facilitate the implementation of patient triage and tiered medical services during epidemic peaks.

17 August 2023 in Machines
Verified
Journal Article
Open access
PDF file
Xiaodong WuXiaodong Wu
Xiaodong Wu
CSChengrui Su
Chengrui Su
LYLiang Yan
Liang Yan
Human–machine shared control of intelligent vehicles is considered an important technology during the industrial application of autonomous driving systems. Among the engineering practices in driver assistance systems, shared steering control is one of the important applications for the human–machine interaction. However, how to deal with human–machine conflicts during emergency scenarios is the main challenge for the controller’s design. Most shared control approaches usually generate machine-oriented results without enough attention to the driver’s reaction. By taking the human driver and machine system as two intelligent agents, this paper proposes a game-based control scheme to achieve a dynamic authority allocation during the lane changing maneuver. Based on the modeling of predicted trajectories of the human driver, a human-intention-based shared steering control is designed to achieve dynamic Nash game equilibrium. Moreover, a human-oriented shared steering mechanism is employed to not only benefit from automated machine assistance, but also make full play of human contributions. Using quantitative comparative analysis in lane changing scenarios with different human–machine conflicts, a better performance by considering both driving comfort and safety is achieved.

23 May 2025 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
LYLiang Yan
Liang Yan
Xiaodong WuXiaodong Wu
Xiaodong Wu
Chongfeng WeiChongfeng Wei
Chongfeng Wei
ZYZhouhang Yu
Zhouhang Yu
HLHangyu Lu
Hangyu Lu
Extreme environments such as icy roads can easily lead to vehicular instability-induced accidents. The insight is that under such operating conditions, the motion of vehicle-tire integrated system shows strong nonlinearity, while driver’s handling experience is insufficient. This article proposes a cooperative driving framework aiming for robust shared control of intelligent vehicles. Specifically, a human-guided shared control (HGSC) strategy is developed to assist human drivers to keep the vehicle controllable under low adhesion conditions. To attain the goal, a hierarchical drift-grip adaptive controller (H-DGAC) is designed out of event-triggered model predictive control problem. Besides, a knowledge-based dual-level Takagi–Sugeno–Kang (DL-TSK) model is established to infer the guidance information injected by the driver. Finally, the proposed strategy is verified by human-in-the-loop experiments, exhibiting its prominent superiority for vehicle stability enhancement.

4 October 2024
Verified
Book Chapter
Open access
PDF file
LYLiang Yan
Liang Yan
Xiaodong WuXiaodong Wu
Xiaodong Wu
HLHangyu Lu
Hangyu Lu
This paper presents a novel human-centered collaborative driving scheme using model-free reinforcement learning (RL) approach. The human-machine cooperation is achieved in both decision-making and steering control levels to improve driving safety while leaving space for human freedom as much as possible. A Markov decision process is firstly derived from the collaborative driving problem, then a RL agent is developed and trained to cooperatively control the vehicle steering under the guidance of a heuristic reward function. Twin delayed deep deterministic policy gradient (TD3) is conducted to attain the optimal control policy. In addition, two extended algorithms with distinct agent action definitions and training patterns are also devised. The effectiveness of the RL-based copilot system is finally validated in an obstacle avoidance scenario by simulation experiments. Driving performance and training efficiency of different RL agents are measured and compared to demonstrate the superiority of the proposed method.

28 January 2026 in Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering
Verified
Journal Article
Closed access
XWXiaodong Wu
Xiaodong Wu
LYLiang Yan
Liang Yan
ZKZhuo Kong
Zhuo Kong
CWChongfeng Wei
Chongfeng Wei
HLHangyu Lu
Hangyu Lu
The advancement in modern control technology makes it possible for intelligent vehicles to go beyond the traditional controllable region and expand the ability to cope with extreme conditions, such as drift with large sideslips. This paper proposes a hierarchical shared drifting control (H-SDC) approach for automated vehicles, aiming to assist inexperienced drivers in completing drift maneuvers, while leaving sufficient control freedom for them. For the cooperative decision layer, the driver intention is integrated into the reference drift equilibrium calculation via two fuzzy inference systems in longitudinal and lateral directions. The operating mode of the collaborative system together with the switching strategy is meticulously defined subsequently. For the real-time execution layer, a robust controller based on time-triggered linear parametric varying-model predictive control (LPV-MPC) is developed for the dynamic drifting state tracking. Efficient implementation of the control algorithm is achieved with the LPV model compensation. Stability limits are specified by an extended stability-oriented safety envelope (SOSE) in the phase plane. The proposed H-SDC framework is verified with both numerical simulation and human-in-the-loop experiment. The results illustrate the effectiveness and superiority of the semi-autonomous system in terms of stability control and drift facilitation. The subjective questionnaire survey further indicates its accessible advantages for novice drivers to learn drifting skills and enjoy the driving pleasure.