Zian Ma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Zian Ma has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Donghao Zhou
Department of Traffic Engineering, Tongji University, 12476 Shanghai, Shanghai China
1 shared publication - Jian Sun
Tongji University
1 shared publication
Latest Publications
- 25 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 5 (2) , 204-216
Autonomous Vehicles’ Turning Motion Planning for Conflict Areas at Mixed-Flow Intersections
An essential task for autonomous vehicles (AVs) is turning at mixed-flow intersections which need to interact with motorized, non-motorized and pedestrian flow simultaneously. In order to handle the multi-interaction environment and uncertainty in perception detection, this paper proposes a hierarchical framework dividing the highly-related driving process into decision, plan and action layers. The decision layer firstly employs logit model combined with Bayes‛ theorem to make discrete choice about whether to turn. Then, the plan layer initializes the local trajectory to the selected waypoints considering the location of the interacting objects by distance potential field. Next, the action layer adjusts the local trajectory and calculates the control parameters of the AV under the impact of the artificial potential field. Finally, feedback is triggered to adjust the decision and trajectory plan when the collision risk increases due to the unexpected behavior of other objects. Furthermore, extend Kalman filter is used to tackle the sensor data noise of AVs. The proposed model is tested at two intersections and the results show the travel time doesn't show statistical significant difference with vehicles that driven by human being. Meanwhile, the AVs have better performance in the aspects of safety and comfort than manually driving.
- 25 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 5 (2) , 204-216
Autonomous Vehicles’ Turning Motion Planning for Conflict Areas at Mixed-Flow Intersections
An essential task for autonomous vehicles (AVs) is turning at mixed-flow intersections which need to interact with motorized, non-motorized and pedestrian flow simultaneously. In order to handle the multi-interaction environment and uncertainty in perception detection, this paper proposes a hierarchical framework dividing the highly-related driving process into decision, plan and action layers. The decision layer firstly employs logit model combined with Bayes‛ theorem to make discrete choice about whether to turn. Then, the plan layer initializes the local trajectory to the selected waypoints considering the location of the interacting objects by distance potential field. Next, the action layer adjusts the local trajectory and calculates the control parameters of the AV under the impact of the artificial potential field. Finally, feedback is triggered to adjust the decision and trajectory plan when the collision risk increases due to the unexpected behavior of other objects. Furthermore, extend Kalman filter is used to tackle the sensor data noise of AVs. The proposed model is tested at two intersections and the results show the travel time doesn't show statistical significant difference with vehicles that driven by human being. Meanwhile, the AVs have better performance in the aspects of safety and comfort than manually driving.