2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)最新文献

{"title":"A Novel Control Strategy of Straight-line Driving Stability for 4WID Electric Vehicles Based on Sliding Mode Control*","authors":"Z. Liu, Yiran Qiao, Xinbo Chen","doi":"10.1109/CVCI54083.2021.9661119","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661119","url":null,"abstract":"Aiming at the stability problem of four-wheel independent drive (4WID) electric vehicles in straight-line driving on low-adhesion roads, a novel sliding mode control strategy is proposed to improve the safety and stability of the vehicle. A 7-degree-of-freedom dynamic model of the vehicle is firstly established and the stability situation is analyzed. According to the analysis results, an additional yaw moment controller based on sliding mode control (SMC) is designed. The yaw angle, yaw rate and lateral displacement are selected as the control variables, and the bilateral adjustment method is adopted to adjust the vehicle attitude in time and prevent vehicle deviation. A slip rate controller based on PID control is also designed to improve the stability margin. Based on the Carsim-Simulink co-simulation platform, the strategy was verified under the conditions of uniform low-adhesion roads and split roads. Simulation results indicate that compared with no control strategy, the proposed strategy has achieved better results and can effectively improve the straight-line driving stability of vehicles.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131368639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Copyright notice]","authors":"","doi":"10.1109/cvci54083.2021.9661203","DOIUrl":"https://doi.org/10.1109/cvci54083.2021.9661203","url":null,"abstract":"","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121286621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning-based voltage degradation prediction with uncertainty quantifications for PEMFC","authors":"Xin Yang, Fengxiang Chen, Jieran Jiao, Shiguang Liu","doi":"10.1109/CVCI54083.2021.9661176","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661176","url":null,"abstract":"The voltage degradation trend prediction has been attached great importance to the application of proton exchange membrane fuel cell systems. Based on Gaussian process regression (GPR) and Bayesian neural network (BNN), this paper uses proton exchange membrane fuel cell dataset to predict the voltage degradation trend. Further, the uncertainty representation of prediction results is realized based on GPR and BNN. Finally, the mean absolute error and root mean square error are used as performance indicators to evaluate the prediction performance. The experimental results show that the evaluation indexes of BNN reach 0.00523, 0.00486 for mean absolute error and 0.007145, 0.007283 for root mean square error, which are less than that of GPR.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127172938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Velocity Prediction Based on LSTM: Impact of Different Input Settings on Prediction Performance","authors":"Jialin Wang, Shiying Dong, Qifang Liu, B. Gao, D. Song","doi":"10.1109/CVCI54083.2021.9661252","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661252","url":null,"abstract":"A large number of intelligent driving systems that rely on the velocity prediction of the host vehicle or other road users are constantly emerging. With the development of nonparametric methods, artificial neural network has been widely employed in the predictive task in the past few years and a significant representative is long short-term neural network (LSTM NN). One of the noteworthy advantages of LSTM is its outstanding ability to overcome the issue of back-propagated error decay and thus demonstrated excellent effect of time series forecasting with long-term dependence. At present LSTM has been deeply researched and has been proved to be a very effective manner to capture nonlinear velocity dynamics. In this paper, various input settings are introduced to study how different variables affect the predictive performance of LSTM. Historical acceleration, the velocity of preceding vehicle, and the distance between adjacent cars are used as supplementary information that input to the model for velocity prediction and the application of real data validated that the predictive performance of the model varies with the input variables. The results show that the inclusion of the velocity of preceding vehicle help to enhance the performance of the model best overall.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"689 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125687405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Generation Method of Synthetic Images with Reduced Domain Gap for Car Detection","authors":"Yubin Huangfu, Weiwen Deng, Bingtao Ren, Juan Ding","doi":"10.1109/CVCI54083.2021.9661221","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661221","url":null,"abstract":"Deep learning has become the main way of the object detection task for autonomous vehicles. Meanwhile, this method typically requires vast amounts of training data to reach their full potential. However, collecting the data from real world and labeling manually is an expensive, time-consuming and error-prone process. Synthetic image has the potential to replace real image for training neural networks, because image creation and labeling annotations are free in this way. For the network trained by synthetic images, the reality gap between real and synthetic images is the main obstacle to use it in the real world. And most previous works are only devoted to generate synthetic images with a good performance on model training, but lack of analysis of the domain gap that affects the performance. This work designs a method of generating the real and synthetic images to analyze and reduce the reality gap between synthetic and real images for car detection. Firstly, this work put one single car with no-background in a random background image to generate real single car and synthetic single car images. In order to further reduce the domain gap in content level, this method keeps the car distribution in synthetic images is similar with the distribution of car in real world. For the purpose of reducing the domain gap in appearance level, the parameters of camera model are same as the camera parameters of image collecting cars and the image is rendered by using the PBRT(Physically Based Ray Tracing) when we generated the synthetic images. Secondly, by training the neural network of instance segmentation with different datasets, the across validation result proves that the reality gap between synthetic and real images is no more than the domain gap between real images. Thirdly, the training results of datasets with different samples diversity show that the diversity of the samples yields better generalization between different datasets for car detection which can effectively reduce the domain gap.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124203268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Anode Pressure Control of Fuel Cell Based on Model Predictive Control","authors":"Fengxiang Chen, Chenghong Lin, Fenglai Pei","doi":"10.1109/CVCI54083.2021.9661137","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661137","url":null,"abstract":"In this paper, a hydrogen supply system model of hydrogen fuel cell system with common rail is established. And then a kind of model predictive control—model algorithm control is taken to control the number of solenoid valves opened in common rail, so as to control the pressure of anode flow field. Finally, the simulation results show that this control method performs well in rise time, overshoot and pressure fluctuation. It means that this control method has characteristics of fast dynamic response and high steady-state precision.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130631677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LTVMPC for Dynamic Positioning of An Autonomous Underwater Vehicle","authors":"Yuheng Chen, Yougang Bian, Qingjia Cui, Lisheng Dong, Biao Xu, Manjiang Hu","doi":"10.1109/CVCI54083.2021.9661196","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661196","url":null,"abstract":"Autonomous Underwater Vehicles (AUVs) need to keep the orientation and position in a dynamic positioning task against ocean currents. Then, a dynamic positioning controller combining linear time-varying model predictive control (LTVMPC) and integral sliding mode control (ISMC) is proposed. The LTVMPC is used to achieve optimal control considering input and output constraints, while the ISMC is adopted to reject external disturbances to improve robustness. Finally, numerical simulations validate the performance of the proposed dynamic positioning controller through a comparison with nonlinear MPC (NMPC) and LTVMPC without ISMC.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128189394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Shared Control Method for Coal Mine Rescue Robots","authors":"Linna Zhou, Yanwei Dou, Hao Liu, Weina Zhang, J. Hu, Chunyu Yang","doi":"10.1109/CVCI54083.2021.9661251","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661251","url":null,"abstract":"This paper studies the man-machine shared control method based on man-machine mutual trust model and environmental information to enhance the working efficiency of coal mine rescue robots. Firstly, the framework of shared control system and working mode for the rescue robots are designed. Secondly, a mutual trust model between humans and machines is established by integrating the status of the operator, the performance of the robot, and the environmental information. Then, a human-machine shared control method is proposed, where the shared control coefficient is reasonably allocated by a fuzzy strategy taking human-machine mutual trust and environmental information as input. Finally, a robot experimental platform is used to illustrate the effectiveness of the proposed shared control method. The results show that the shared control method is of higher efficiency than teleoperation operation and autonomous control methods.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132038959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lane Departure Risk Assessment for Hands-free Driving Functions","authors":"Daofei Li, Bin Xiao, Siyuan Lin","doi":"10.1109/CVCI54083.2021.9661247","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661247","url":null,"abstract":"Hands-free driving has become increasingly appealing and popular as an advanced form of SAE Level 2 automation, especially in premium brands. Such kind of advanced driver assistance system must ensure safety via a risk assessment module to initiate human take-over request if necessary. In lane keeping scenario, accurate trajectory prediction of the ego vehicle is vital to lane departure risk assessment. Thanks to automated driving, the actual control laws are known or can be learnt, which can support more precise prediction. Here we propose a Kalman predictor with actual control laws for future ego vehicle trajectory prediction. With a simulated scenario with varying velocity and road curvature, the algorithm is proved effective and outperforms traditional physics-based trajectory prediction benchmarks. Comparison between algorithms considering only lateral control law is also carried out, and results show that the algorithm considering both longitudinal and lateral control laws has better prediction accuracy. The proposed algorithm is promising to be applied in risk assessment module of hands-free driving functions.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129007804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Path planning for automatic parking based on improved Hybrid A* algorithm","authors":"Lu Xiong, Jie Ying Gao, Zhiqiang Fu, Kui Xiao","doi":"10.1109/CVCI54083.2021.9661197","DOIUrl":"https://doi.org/10.1109/CVCI54083.2021.9661197","url":null,"abstract":"In this paper, an automatic parking planning algorithm based on hybrid A* improvement is proposed for the unstructured environment of parking lots. In order to obtain an optimal trajectory under the conditions of safety and comfort, the algorithm makes the following adjustments: (1) Based on the kinematic and dynamical parameters of the experimental vehicle, the expansion search of the nodes is performed using clothoids to ensure the continuity of curvature. (2) The heuristic term of hybrid A* is improved by adding a penalty term of obstacle distance to the Reeds-shepp (RS) curve as the heuristic term, which improves the problem that the RS curve is too close to the obstacle. The effectiveness of the algorithm is evaluated based on simulation and real-world experimental results, and it is demonstrated that the algorithm can be widely used in automatic parking scenarios.","PeriodicalId":419836,"journal":{"name":"2021 5th CAA International Conference on Vehicular Control and Intelligence (CVCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128779450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}