-
1.发明申请公开(公告)号：US20240375682A1
公开(公告)日：2024-11-14
申请号：US18259378
申请日：2022-02-25
Applicant: Southeast University
Inventor： Xu LI , Weiming HU , Jinchao HU , Kun WEI , Qimin XU
Abstract: A method of making a highly humanoid safe driving decision for an automated driving commercial vehicle, includes: collecting synchronously multi-source information on driving behaviors in typical traffic scenarios, constructing an expert trajectory data set representing driving behaviors of excellent drivers; simulating the driving behaviors of excellent drivers by utilizing a generative adversarial imitation learning (GAIL) algorithm, in a comprehensive consideration of influences of factors such as a forward collision, a backward collision, a transverse collision, a vehicle roll stability and a driving smoothness on a driving safety, constructing a generator and a discriminator by utilizing a proximal policy optimization algorithm and a deep neural network respectively, and establishing a safe driving decision-making model with highly humanoid level; and training the safe driving decision-making model to obtain safe driving policies under different driving conditions and to implement an output of an advanced decision-making for the automated driving commercial vehicle.
-
公开(公告)号：US20230013071A1
公开(公告)日：2023-01-19
申请号：US17784806
申请日：2021-09-23
Applicant: Southeast University
IPC: G01M17/007 , G07C5/00 , G08G1/01
Abstract: Disclosed are a Type-II Autonomous Emergency Braking System (AEBS) test and evaluation device and method based on a BeiDou space-time reference, where the device includes three parts: a roadside-end information acquisition module, a vehicle-end information acquisition module, and an integrated information processing module. The roadside-end information acquisition module can acquire accurate message sending time by means of a BeiDou time service unit; the vehicle-end information acquisition module can acquire accurate time of receiving a roadside-end message, information acquired by a combined inertial navigation unit, and audio/vibration information acquired by a Single Chip Microcomputer (SCM) embedded unit; and the integrated information processing module can implement accurate, quantitative test and evaluation of indexes such as a vehicle-road communication delay and warning signal sending time. The method of the present disclosure performs data analysis and processing based on a globally unified BeiDou space-time reference and by means of a Support Vector Machine (SVM)-based dynamic Hermite interpolation method, which has an accurate test and evaluation result. Further, the method does not have any requirements for a communication system of the type-II AEBS, thus achieving convenient testing and a wide range of application.
-
公开(公告)号：US20220348174A1
公开(公告)日：2022-11-03
申请号：US17764871
申请日：2021-04-12
Applicant: SOUTHEAST UNIVERSITY
IPC: B60T8/1755
Abstract: For a tank truck using an EBS, the present invention provides a tank truck rollover relieved control method based on electronic braking deceleration. Firstly, a tank truck rollover scene applicable to the relieved control method is defined; then, a least square method is adopted to establish a characterization function of tank truck braking deceleration; and finally, tank truck rollover relieved control is achieved on the basis of the characterization function of the braking deceleration and the EBS. The method fits out a function expression of the tank truck braking deceleration and can automatically select a proper braking deceleration under different rollover scenes according to kinematics information of the tank truck and vehicle body information; during tank truck braking deceleration, an operation of a driver is considered, so that man-machine effective combination is achieved; and relieved braking deceleration is conducted when the tank truck is in a potential rollover risk state, a situation that emergency braking is conducted when the tank truck has high rollover risk is avoided, and tank truck rollover control stability and effectiveness are improved.
-
公开(公告)号：US20220179061A1
公开(公告)日：2022-06-09
申请号：US17436614
申请日：2019-05-29
Applicant: Southeast University
Inventor： Xu LI , Huaikun GAO , Qimin XU
Abstract: In Intelligent Vehicle Infrastructure Cooperative Systems (IVICS), a high-precision vehicle positioning method utilizing Ultra-Wide Band (UWB) is proposed. Owing to remarkable wide band of radio signal, this UWB-based positioning method shows excellent anti-interference capability and multi-path immunity, which are essential for achieving high precision in practical traffic scenario. In this approach, several UWB nodes are deliberately deployed at the crossing with the help of roadside infrastructure in IVICS. Meanwhile, an algorithm aiming at Non Line of Sight (NLOS) error compensation is developed to improve the positioning performance. In a word, this method has been demonstrated the potential to achieve accurate, reliable, continuous and integrated localization.
-
Patent Agency Ranking
公开(公告)号：US20230410659A1
公开(公告)日：2023-12-21
申请号：US18249715
申请日：2021-04-12
Applicant: SOUTHEAST UNIVERSITY
Inventor： Xu LI , Jinchao HU , Weiming HU
IPC: G08G1/16 , G06V20/70 , G06V10/82 , G06V10/26 , G06V40/20 , G06V10/77 , G06V10/774 , G06V20/52 , G08G1/052
CPC classification number: G08G1/166 , G06V20/70 , G06V10/82 , G06V10/26 , G06V40/25 , G06V10/7715 , G06V10/774 , G06V20/52 , G08G1/052
Abstract: A method for predicting a pedestrian crossing behavior for an intersection includes the following steps: step 1: designing an immediate reward function; step 2: establishing a fully convolutional neural network-long-short term memory network (FCN-LSTM) model to predict a motion reward function; step 3: training the fully convolutional neural network-long-short term memory network (FCN-LSTM) model based on reinforcement learning; and step 4: predicting the pedestrian crossing behavior and performing hazard early-warning. The technical solution does not require establishment of a complex pedestrian movement model or preparation of massive labeled data sets, achieves autonomous learning of pedestrian crossing behavior features at the intersection, predicts their walking, stopping, running and other behaviors, especially predicts the pedestrian crossing behavior when inducing hazards such as pedestrian-vehicle collision and scratch in real time, and performs hazard early-warning on crossing pedestrians and passing vehicles.
-
公开(公告)号：US20210312653A1
公开(公告)日：2021-10-07
申请号：US17057915
申请日：2019-05-29
Applicant: SOUTHEAST UNIVERSITY
Inventor： Xu LI , Jiwen CAO , Peizhou NI , Kun WEI
Abstract: The present invention discloses a positive azimuth towing guidance method for road rescue equipment based on license plate corner features. The method combines a structure of the road rescue equipment and characteristics of a positive azimuth towing operation. First, an image of an operation area is collected by installing a camera, and grayscale processing and Gaussian smooth filtering are performed on the image; corner detection is performed on the smoothed grayscale image, and preference is implemented according to corner strengths; hierarchical clustering is performed on the preferred corners; an effective corner set of license plate characters is sorted out to implement license plate locating; and then towing guidance is implemented according to a license plate locating result, to improve the rescue efficiency of the road rescue equipment. The guidance method provided in the present invention has good real-time performance, environmental adaptability and anti-interference ability, thereby effectively improving the rescue efficiency of the road rescue equipment.
-
公开(公告)号：US20210303911A1
公开(公告)日：2021-09-30
申请号：US17267493
申请日：2019-05-16
Applicant: SOUTHEAST UNIVERSITY
Inventor： Xu LI , Zhiyong Zheng , Kun Wei
Abstract: The present invention discloses a roadside image pedestrian segmentation method based on a variable-scale multi-feature fusion convolutional network. For scenes where the pedestrian scale changes significantly in the intelligent roadside terminal image, this method designs two parallel convolutional neural networks to extract the local and global features of pedestrians at different scales in the image, and then fuses the local features and global features extracted by the first network with the local features and global features extracted by the second network at the same level, and then fuse the fused local features and global features for the second time to obtain a variable-scale multi-feature fusion convolutional neural network, and then train the network and input roadside pedestrian images to realize pedestrian segmentation. The present invention effectively solves the problems that most current pedestrian segmentation methods based on a single network structure are prone to segmentation boundary fuzziness and missing segmentation.
-
公开(公告)号：US20230182725A1
公开(公告)日：2023-06-15
申请号：US17766870
申请日：2021-04-12
Applicant: SOUTHEAST UNIVERSITY
Inventor： Xu LI , Weiming HU , Jinchao HU , Xuefen ZHU
IPC: B60W30/095 , G06N3/045 , G06N3/0464
CPC classification number: B60W30/0956 , G06N3/045 , G06N3/0464 , B60W2554/4041 , B60W2420/52 , B60W2554/80 , B60W2300/125
Abstract: The present invention discloses a backward anti-collision driving decision-making method for a heavy commercial vehicle. Firstly, a traffic environment model is established, and movement state information of a heavy commercial vehicle and a vehicle behind the heavy commercial vehicle is collected. Secondly, a backward collision risk assessment model based on backward distance collision time is established, and a backward collision risk is accurately quantified. Finally, a backward anti-collision driving decision-making problem is described as a Markov decision-making process under a certain reward function, a backward anti-collision driving decision-making model based on deep reinforcement learning is established, and an effective, reliable and adaptive backward anti-collision driving decision-making policy is obtained. The method provided by the present invention can overcome the defect of lack for research on the backward anti-collision driving decision-making policy for the heavy commercial vehicle in the existing method, can quantitatively output proper steering wheel angle and throttle opening control quantities, can provide effective and reliable backward anti-collision driving suggestions for a driver, and can reduce backward collision accidents.
-
公开(公告)号：US20220379893A1
公开(公告)日：2022-12-01
申请号：US17768765
申请日：2021-04-12
Applicant: SOUTHEAST UNIVERSITY
Inventor： Xu LI , Weiming HU , Qimin XU , Kun WEI
IPC: B60W30/18 , B60W30/165 , B60W40/114 , B60W40/13 , B60W40/072 , B60W50/00 , G08G1/00 , G05D1/02
Abstract: The present invention discloses an intelligent vehicle platoon lane change performance evaluation method. First, an intelligent vehicle platoon lane change performance test scenario is established; secondly, a three-degree of freedom nonlinear dynamics model is established according to motion characteristics of intelligent vehicles in a platoon lane change process; further, an improved adaptive unscented Kalman filter algorithm is utilized to perform filter estimation on state variables of positions and velocities of platoon vehicles; and finally, based on accurately recursive vehicle motion state parameters, evaluation indexes for platoon lane change performance are proposed and quantified, and an evaluation system for platoon lane change performance is constructed. According to the method proposed in the present invention, the problem of lacking platoon lane change performance quantitative evaluation at present is solved, vehicle motion state parameters can be measured in a high-precision and comprehensive manner, multi-dimensional platoon lane change performance evaluation indexes are quantified and output, and comprehensive, accurate, and reliable scientific quantitative evaluation for platoon lane change performance is achieved.
-
-
-
-
-
-
-
-
Search Results
9
records
(took 0.017 seconds)
