陈慧岩
作者: Yuxuan Han1;Qingxiao Liu1;Haiou Liu1;Boyang Wang1;Zheng Zang1;Huiyan Chen1; (1College of Mechanical Engineering, Beijing Institute of Technology, Beijing, China)
出处: IEEE Transactions on Intelligent Vehicles 2024 Vol.9 No.1 P2210-2222
关键词: Trajectory;Transformers;Adaptation models;Predictive models;Semantics;Kinematics;Vehicle dynamics;Trajectory Prediction;Learning-based Framework;Trajectory Prediction Method;Low Efficiency;Urban Environments;Attention Mechanism;Graph Convolutional Network;Prediction Framework;Presence Of Adaptation;Central Agent;Time Step;Local Area;Multilayer Perceptron;Learning-based Methods;Number Of Agents;Future Trajectories;Historical Trajectory;Number Of Heads;Rule-based Methods;Neighboring Agents;Multilayer Perceptron Network;Prediction Scenarios;Multi-head Self-attention;Trajectories Of Agents;Displacement Error;Graph Attention Network;Attention Heads;Temporal Complexity
摘要: Effective trajectory prediction is of great significance for the design of intelligent driving systems. To overcome the problems of low algorithm effi ...
作者: QingXiao Liu1;Hui Yao1;Chao Lu1;HaiOu Liu1;Yangtian Yi1;HuiYan Chen1; (1School of Mechanical Engineering, Beijing Institute of Technology, Beijing, China)
出处: IEEE Transactions on Industrial Electronics 2024 Vol.71 No.6 P6396-6406
关键词: Visualization;Predictive models;Vehicles;Semantics;External stimuli;Cognition;Vocabulary
摘要: An object-level attention prediction framework for drivers in the urban environment with rich semantic and motion information is proposed in this arti ...
作者: Zhou, Wei1; Zhang, Ruizeng2; Li, Leida3; Yue, Guanghui4; Gong, Jianwei2; Chen, Huiyan2; Liu, Hantao1 (1School of Computer Science and Informatics, Cardiff University, Cardiff, U.K;2School of Mechanical Engineering, Beijing Institute of Technology, Beijing, China;3School of Artificial Intelligence, Xidian University, Xi'an, China;4School of Biomedical Engineering, Shenzhen University, Shenzhen, China)
出处: IEEE Transactions on Intelligent Vehicles 2024 P1-16
作者: Xiang Zhang1;Boyang Wang1;Yaomin Lu1;Haiou Liu1;Jianwei Gong1;Huiyan Chen1; (1School of Mechanical Engineering, Beijing Institute of Technology, Beijing, China)
出处: IEEE Transactions on Intelligent Vehicles 2024 Vol.9 No.1 P2675-2687
关键词: Planning;Safety;Intelligent vehicles;Trajectory optimization;Vehicle dynamics;Reinforcement learning;Programming;Interactive;Path Planning;Planning Methods;Nonlinear Programming;Quadratic Programming;Configuration Space;Trajectory Optimization;Vehicle Dynamics;General Scenario;Trajectory Optimization Problem;Experimental Group;Deep Learning;Unit Time;Maximum Speed;Long Short-term Memory;Reason For This Phenomenon;Deep Reinforcement Learning;Formation Control;Vehicle Position;Hierarchical Framework;Nonlinear Programming Problem;Final Trajectory;Integer Variables;Trajectory Generation;Traffic Scenarios;Vehicle Weight;Smooth Trajectory;Vehicle Trajectory;Maximum Acceleration;Trajectory Planning
摘要: Multi-vehicle coordinated motion planning has always been challenged to safely and efficiently resolve conflicts under non-holonomic dynamic constrain ...
作者: Chen Shuping1;,Chen Huiyan2;,Zhao Zhiguo1; (1Tongji University;2Beijing Institute of Technology)
出处: 2023中国汽车工程学会年会暨展览会 2023
会议录: 2023中国汽车工程学会年会论文集(1)
关键词: autonomous vehicles;trajectory tracking;MPC;yaw stability;rollover prevention
摘要: Most researches concerning the yaw stability and rollover prevention for autonomous vehicles are studied separately and decoupled the longitudinal and ...
发明人: 刘海鸥,李世豪,薛明轩,张翔,李志伟,王博洋,陈慧岩
申请人: 北京理工大学
申请号: 202311088086.3
申请日期: 2023.08.28
摘要: 本发明涉及一种无人越野车辆通过障碍地形的安全行为检测方法,属于车辆安全控制技术领域,解决了现有障碍地形的路径可通行性不高的问题。包括:获取规划路线上未检测的轨迹点序列,从轨迹点序列中获取障碍地形内的轨迹点作为待检测轨迹点;根据地面模型计算出无人越野车辆在待检测轨迹点的位姿和各车轮接触点数据;根据各车 ...
发明人: 刘海鸥,薛明轩,王博洋,陈慧岩
申请人: 北京理工大学
申请号: 202311198777.9
申请日期: 2023.09.18
摘要: 本发明涉及一种地面无人装备自主行进间射击的安全行为检测方法及系统,属于车辆安全控制技术领域,解决了现有缺少自主行进间射击的安全性检测问题。包括:根据各轨迹点的复合姿态的变尺度包络,将存在环境点云的数量小于等于阈值的变尺度包络对应的轨迹点的第一标记设置为安全,否则设置为风险;对风险轨迹点获取地面无人装 ...
发明人: 刘海鸥,李世豪,薛明轩,张翔,李志伟,王博洋,陈慧岩
申请人: 北京理工大学
申请号: 202311088074.0
申请日期: 2023.08.28
摘要: 本发明涉及一种无人越野车辆自主行驶的安全行为检测方法,属于车辆安全控制技术领域,解决了现有未考虑复杂地形和路面条件导致安全检测不足的问题。包括:获取规划路线上未检测的轨迹点序列;根据地面模型计算出各轨迹点的位姿和接触点数据;根据位姿依次检测无人越野车辆在各轨迹点是否满足静态位姿约束和动态稳定约束,如 ...
发明人: 刘海鸥,陶俊峰,关海杰,陈慧岩
申请人: 北京理工大学
申请号: 202310395181.1
申请日期: 2023.04.14
摘要: 本发明涉及一种面向越野环境的无人驾驶车辆局部路径规划方法,属于路径规划技术领域,解决了现有局部路径的规划忽略了地形特性而导致路径可行性不高的问题。包括基于分层地图,根据曲率和单位攀升高度,获取全局参考路径的纵向采样点,得到多个纵向采样阶段;根据纵向采样阶段的长度和预置的横偏量,对阶段终点的纵向采样点 ...
发明人: 谭颖琦,陈慧岩,龚建伟,臧政
申请人: 北京理工大学,北京工业职业技术学院
申请号: 202311514751.0
申请日期: 2023.11.15
摘要: 本发明公开一种串联混合动力无人履带车辆的能量管理方法及系统,涉及能量管理技术领域,基于串联混合动力无人履带车辆的多目标奖励函数对基于深度确定性策略梯度算法的能量管理控制器进行训练,得到训练好的能量管理控制器,在实际应用时,直接以串联混合动力无人履带车辆的车辆参数为输入,利用训练好的能量管理控制器即可 ...