his paper adopts the 3-3-2 information processing method for the capture of moving objects as its premise, and proposes a basic principle of three-dimensional (3D) imaging using biological compound eye. Traditional bi...his paper adopts the 3-3-2 information processing method for the capture of moving objects as its premise, and proposes a basic principle of three-dimensional (3D) imaging using biological compound eye. Traditional bionic vision is limited by the available hardware. Therefore, in this paper, the new-generation technology of microlens-array light-field camera is proposed as a potential method for the extraction of depth information from a single image. A significant characteristic of light-field imaging is that it records intensity and directional information from the lights entering the camera. Herein, a refocusing method using light-field image is proposed. By calculating the focusing cost at different depths from the object, the imaging plane of the object is determined, and a depth map is constructed based on the position of the object’s imaging plane. Compared with traditional light-field depth estimation, the depth map calculated by this method can significantly improve resolution and does not depend on the number of light-field microlenses. In addition, considering that software algorithms rely on hardware structure, this study develops an imaging hardware that is only 7 cm long based on the second-generation microlens camera’s structure, further validating its important refocusing characteristics. It thereby provides a technical foundation for 3D imaging with a single camera.展开更多
In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucia...In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucial for the vehicle design and safe control.However,precise state estimation of the brake pressure is desired to perform safe driving with a high degree of autonomy.In this paper,a sensorless state estimation technique of the vehicle's brake pressure is developed using a deep-learning approach.A deep neural network(DNN)is structured and trained using deep-learning training techniques,such as,dropout and rectified units.These techniques are utilized to obtain more accurate model for brake pressure state estimation applications.The proposed model is trained using real experimental training data which were collected via conducting real vehicle testing.The vehicle was attached to a chassis dynamometer while the brake pressure data were collected under random driving cycles.Based on these experimental data,the DNN is trained and the performance of the proposed state estimation approach is validated accordingly.The results demonstrate high-accuracy brake pressure state estimation with RMSE of 0.048 MPa.展开更多
Dear Editor, With the development of automobile industry and artificial intelligence(AI) domains, autonomous vehicles(AVs) are becoming a reality and promise to revolutionize human mobility [1]–[3]. The decision-maki...Dear Editor, With the development of automobile industry and artificial intelligence(AI) domains, autonomous vehicles(AVs) are becoming a reality and promise to revolutionize human mobility [1]–[3]. The decision-making system of AVs is crucial, which is typically required to trade off multiple competing objectives. For example,when determining driving policies.展开更多
Improving the capacity of intersections is the key to enhancing road traffic systems.Benefiting from the application of Connected Automated Vehicles(CAVs)in the foreseeing future,it is promising to fully utilize spati...Improving the capacity of intersections is the key to enhancing road traffic systems.Benefiting from the application of Connected Automated Vehicles(CAVs)in the foreseeing future,it is promising to fully utilize spatiotemporal resources at intersections through cooperative and intelligent trajectory planning for CAVs.Lane-free traffic is currently a highly anticipated solution that can achieve more flexible trajectories without being limited by lane boundaries.However,it is challenging to apply efficient lane-free traffic to be compatible with the traditional intersection control mode for mixed flow composed of CAVs and Human-driving Vehicles(HVs).To address the research gap,this paper proposes a spatiotemporal-restricted A∗algorithm to obtain efficient and flexible lane-free trajectories for CAVs.First,we restrict the feasible area of the heuristic search algorithm by considering the feasible area and orientation of vehicles to maintain the trajectory directionality of different turning behaviors.Second,we propose a spatiotemporal sparse sampling method by defining the four-dimensional spatiotemporal grid to accelerate the execution of the heuristic search algorithm.Third,we consider the motions of HVs as dynamic obstacles with rational trajectory fluctuation during the process of trajectory planning for CAVs.The proposed method can retain the advantage of efficiently exploring feasible trajectories through the hybrid A*algorithm,while also utilizing multiple spatiotemporal constraints to accelerate solution efficiency.The experimental results of the simulated and real scenarios with mixed flows show that the proposed model can continuously enhance traffic efficiency and fuel economy as the penetration of CAVs gradually increases.展开更多
基金The National Major Project Research and Development Project (2017YFB0503003)The National Natural Science Foundation of China(61101157, 60602042).
文摘his paper adopts the 3-3-2 information processing method for the capture of moving objects as its premise, and proposes a basic principle of three-dimensional (3D) imaging using biological compound eye. Traditional bionic vision is limited by the available hardware. Therefore, in this paper, the new-generation technology of microlens-array light-field camera is proposed as a potential method for the extraction of depth information from a single image. A significant characteristic of light-field imaging is that it records intensity and directional information from the lights entering the camera. Herein, a refocusing method using light-field image is proposed. By calculating the focusing cost at different depths from the object, the imaging plane of the object is determined, and a depth map is constructed based on the position of the object’s imaging plane. Compared with traditional light-field depth estimation, the depth map calculated by this method can significantly improve resolution and does not depend on the number of light-field microlenses. In addition, considering that software algorithms rely on hardware structure, this study develops an imaging hardware that is only 7 cm long based on the second-generation microlens camera’s structure, further validating its important refocusing characteristics. It thereby provides a technical foundation for 3D imaging with a single camera.
文摘In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucial for the vehicle design and safe control.However,precise state estimation of the brake pressure is desired to perform safe driving with a high degree of autonomy.In this paper,a sensorless state estimation technique of the vehicle's brake pressure is developed using a deep-learning approach.A deep neural network(DNN)is structured and trained using deep-learning training techniques,such as,dropout and rectified units.These techniques are utilized to obtain more accurate model for brake pressure state estimation applications.The proposed model is trained using real experimental training data which were collected via conducting real vehicle testing.The vehicle was attached to a chassis dynamometer while the brake pressure data were collected under random driving cycles.Based on these experimental data,the DNN is trained and the performance of the proposed state estimation approach is validated accordingly.The results demonstrate high-accuracy brake pressure state estimation with RMSE of 0.048 MPa.
文摘Dear Editor, With the development of automobile industry and artificial intelligence(AI) domains, autonomous vehicles(AVs) are becoming a reality and promise to revolutionize human mobility [1]–[3]. The decision-making system of AVs is crucial, which is typically required to trade off multiple competing objectives. For example,when determining driving policies.
基金This work was supported by the Science and Technology Commission of Shanghai Municipality(Nos.22YF1461400 and 22DZ1100102)the National Key R&D Program of China(No.2022ZD0160104).
文摘Improving the capacity of intersections is the key to enhancing road traffic systems.Benefiting from the application of Connected Automated Vehicles(CAVs)in the foreseeing future,it is promising to fully utilize spatiotemporal resources at intersections through cooperative and intelligent trajectory planning for CAVs.Lane-free traffic is currently a highly anticipated solution that can achieve more flexible trajectories without being limited by lane boundaries.However,it is challenging to apply efficient lane-free traffic to be compatible with the traditional intersection control mode for mixed flow composed of CAVs and Human-driving Vehicles(HVs).To address the research gap,this paper proposes a spatiotemporal-restricted A∗algorithm to obtain efficient and flexible lane-free trajectories for CAVs.First,we restrict the feasible area of the heuristic search algorithm by considering the feasible area and orientation of vehicles to maintain the trajectory directionality of different turning behaviors.Second,we propose a spatiotemporal sparse sampling method by defining the four-dimensional spatiotemporal grid to accelerate the execution of the heuristic search algorithm.Third,we consider the motions of HVs as dynamic obstacles with rational trajectory fluctuation during the process of trajectory planning for CAVs.The proposed method can retain the advantage of efficiently exploring feasible trajectories through the hybrid A*algorithm,while also utilizing multiple spatiotemporal constraints to accelerate solution efficiency.The experimental results of the simulated and real scenarios with mixed flows show that the proposed model can continuously enhance traffic efficiency and fuel economy as the penetration of CAVs gradually increases.
