针对传统人工势场法在多障碍物复杂环境的全局路径规划中出现的目标不可达、易陷入陷阱区域以及局部极小点问题,提出一种简化障碍物预测碰撞人工势场法(simplified obstacles and predict collision of artificial potential field meth...针对传统人工势场法在多障碍物复杂环境的全局路径规划中出现的目标不可达、易陷入陷阱区域以及局部极小点问题,提出一种简化障碍物预测碰撞人工势场法(simplified obstacles and predict collision of artificial potential field method,SOPC-APF),算法引入预测碰撞思想,在机器人未进入陷阱区域或者极小点问题前做出决策;对于多障碍物的斥力与目标点的引力产生的合力使机器人陷入震荡,提出简化障碍物,即简化为影响范围内目标点一侧的受限障碍物;针对目标不可达问题,在碰撞预测基础上,设定虚拟目标点,经改进的斥力函数引导机器人快速生成一条平滑、平稳、无碰撞的路径。通过与传统算法、改进APF算法以及改进蚁群算法的仿真对比实验表明,SOPC-APF有效解决了人工势场法不适用于多障碍物复杂环境的问题,以及传统算法容易陷入陷阱区域和局部极小点问题。展开更多
Risk assessment is a crucial component of collision warning and avoidance systems for intelligent vehicles.Reachability-based formal approaches have been developed to ensure driving safety to accurately detect potenti...Risk assessment is a crucial component of collision warning and avoidance systems for intelligent vehicles.Reachability-based formal approaches have been developed to ensure driving safety to accurately detect potential vehicle collisions.However,they suffer from over-conservatism,potentially resulting in false–positive risk events in complicated real-world applications.In this paper,we combine two reachability analysis techniques,a backward reachable set(BRS)and a stochastic forward reachable set(FRS),and propose an integrated probabilistic collision–detection framework for highway driving.Within this framework,we can first use a BRS to formally check whether a two-vehicle interaction is safe;otherwise,a prediction-based stochastic FRS is employed to estimate the collision probability at each future time step.Thus,the framework can not only identify non-risky events with guaranteed safety but also provide accurate collision risk estimation in safety-critical events.To construct the stochastic FRS,we develop a neural network-based acceleration model for surrounding vehicles and further incorporate a confidence-aware dynamic belief to improve the prediction accuracy.Extensive experiments were conducted to validate the performance of the acceleration prediction model based on naturalistic highway driving data.The efficiency and effectiveness of the framework with infused confidence beliefs were tested in both naturalistic and simulated highway scenarios.The proposed risk assessment framework is promising for real-world applications.展开更多
为实现急弯路段的追尾碰撞风险主动防控,提出了一种基于多源数据融合的追尾冲突动态预测方法。首先,基于无人机、毫米波雷达等采集的车辆运行数据,提出了适用于急弯路段交通流特征的追尾冲突判别模型及冲突等级阈值划分标准,分析了急弯...为实现急弯路段的追尾碰撞风险主动防控,提出了一种基于多源数据融合的追尾冲突动态预测方法。首先,基于无人机、毫米波雷达等采集的车辆运行数据,提出了适用于急弯路段交通流特征的追尾冲突判别模型及冲突等级阈值划分标准,分析了急弯路段的追尾冲突空间分布特征。然后,筛选车型、大车比率、断面速度差等13个交通流特征指标作为输入变量,以粒子群算法为基础,分别构建了其与BP神经网络、随机森林、支持向量机算法的追尾冲突动态组合预测模型,并根据混淆矩阵和曲线下面积评估各模型的预测性能,利用黑箱解释方法分析冲突发生概率的显著性影响因素及影响程度。结果表明:相较于平直或一般弯道路段,急弯路段的追尾冲突TTC(Time to Collision)值更小,出弯缓和曲线段冲突更为严重,且弯道内侧碰撞风险最高;粒子群-随机森林模型的追尾冲突预测性能最佳,灵敏度达90.70%;急弯路段追尾冲突受车辆平均车头间距的影响程度最大,当平均车头间距为25 m左右时,冲突发生概率最小,向心加速度均值、速度均值等因素亦对其有显著影响。展开更多
The purpose of this paper is to alleviate the potential safety problems associated with the human driver and the automatic system competing for the right of way due to different objectives by mitigating the human-mach...The purpose of this paper is to alleviate the potential safety problems associated with the human driver and the automatic system competing for the right of way due to different objectives by mitigating the human-machine conflict phenomenon in human-machine shared driving(HMSD)technology from the automation system.Firstly,a basic lane-changing trajectory algorithm based on the quintic polynomial in the Frenet coordinate system is developed.Then,in order to make the planned trajectory close to human behavior,naturalistic driving data is collected,based on which some lane-changing performance features are selected and analyzed.There are three aspects have been taken into consideration for the human-like lane-changing trajectory:vehicle dynamic stability performance,driving cost optimization,and collision avoidance.Finally,the HMSD experiments are conducted with the driving simulator to test the potential of the human-like lane-changing trajectory planning algorithm.The results demonstrate that the lane-changing trajectory planning algorithm with the highest degree of personalization is highly consistent with human driver behavior and consequently would potentially mitigate the human-machine conflict with the HMSD application.Furthermore,it could be further employed as an empirical trajectory prediction result.The algorithm employs the distribution state of the historical trajectory for human-like processing,simplifying the operational process and ensuring the credibility,integrity,and interpretability of the results.Moreover,in terms of optimization processing,the form of optimization search followed by collision avoidance detection is adopted to in principle reduce the calculation difficulty.Additionally,a new convex polygon collision detection method,namely the vertex embedding method,is proposed for collision avoidance detection.展开更多
文摘针对传统人工势场法在多障碍物复杂环境的全局路径规划中出现的目标不可达、易陷入陷阱区域以及局部极小点问题,提出一种简化障碍物预测碰撞人工势场法(simplified obstacles and predict collision of artificial potential field method,SOPC-APF),算法引入预测碰撞思想,在机器人未进入陷阱区域或者极小点问题前做出决策;对于多障碍物的斥力与目标点的引力产生的合力使机器人陷入震荡,提出简化障碍物,即简化为影响范围内目标点一侧的受限障碍物;针对目标不可达问题,在碰撞预测基础上,设定虚拟目标点,经改进的斥力函数引导机器人快速生成一条平滑、平稳、无碰撞的路径。通过与传统算法、改进APF算法以及改进蚁群算法的仿真对比实验表明,SOPC-APF有效解决了人工势场法不适用于多障碍物复杂环境的问题,以及传统算法容易陷入陷阱区域和局部极小点问题。
基金supported by the proactive SAFEty systems and tools for a constantly UPgrading road environment(SAFE-UP)projectfunding from the European Union’s Horizon 2020 Research and Innovation Program(861570)。
文摘Risk assessment is a crucial component of collision warning and avoidance systems for intelligent vehicles.Reachability-based formal approaches have been developed to ensure driving safety to accurately detect potential vehicle collisions.However,they suffer from over-conservatism,potentially resulting in false–positive risk events in complicated real-world applications.In this paper,we combine two reachability analysis techniques,a backward reachable set(BRS)and a stochastic forward reachable set(FRS),and propose an integrated probabilistic collision–detection framework for highway driving.Within this framework,we can first use a BRS to formally check whether a two-vehicle interaction is safe;otherwise,a prediction-based stochastic FRS is employed to estimate the collision probability at each future time step.Thus,the framework can not only identify non-risky events with guaranteed safety but also provide accurate collision risk estimation in safety-critical events.To construct the stochastic FRS,we develop a neural network-based acceleration model for surrounding vehicles and further incorporate a confidence-aware dynamic belief to improve the prediction accuracy.Extensive experiments were conducted to validate the performance of the acceleration prediction model based on naturalistic highway driving data.The efficiency and effectiveness of the framework with infused confidence beliefs were tested in both naturalistic and simulated highway scenarios.The proposed risk assessment framework is promising for real-world applications.
文摘为实现急弯路段的追尾碰撞风险主动防控,提出了一种基于多源数据融合的追尾冲突动态预测方法。首先,基于无人机、毫米波雷达等采集的车辆运行数据,提出了适用于急弯路段交通流特征的追尾冲突判别模型及冲突等级阈值划分标准,分析了急弯路段的追尾冲突空间分布特征。然后,筛选车型、大车比率、断面速度差等13个交通流特征指标作为输入变量,以粒子群算法为基础,分别构建了其与BP神经网络、随机森林、支持向量机算法的追尾冲突动态组合预测模型,并根据混淆矩阵和曲线下面积评估各模型的预测性能,利用黑箱解释方法分析冲突发生概率的显著性影响因素及影响程度。结果表明:相较于平直或一般弯道路段,急弯路段的追尾冲突TTC(Time to Collision)值更小,出弯缓和曲线段冲突更为严重,且弯道内侧碰撞风险最高;粒子群-随机森林模型的追尾冲突预测性能最佳,灵敏度达90.70%;急弯路段追尾冲突受车辆平均车头间距的影响程度最大,当平均车头间距为25 m左右时,冲突发生概率最小,向心加速度均值、速度均值等因素亦对其有显著影响。
基金Open Fund of State Key Laboratory of Automobile Simulation and Control of Jilin University(20201111).
文摘The purpose of this paper is to alleviate the potential safety problems associated with the human driver and the automatic system competing for the right of way due to different objectives by mitigating the human-machine conflict phenomenon in human-machine shared driving(HMSD)technology from the automation system.Firstly,a basic lane-changing trajectory algorithm based on the quintic polynomial in the Frenet coordinate system is developed.Then,in order to make the planned trajectory close to human behavior,naturalistic driving data is collected,based on which some lane-changing performance features are selected and analyzed.There are three aspects have been taken into consideration for the human-like lane-changing trajectory:vehicle dynamic stability performance,driving cost optimization,and collision avoidance.Finally,the HMSD experiments are conducted with the driving simulator to test the potential of the human-like lane-changing trajectory planning algorithm.The results demonstrate that the lane-changing trajectory planning algorithm with the highest degree of personalization is highly consistent with human driver behavior and consequently would potentially mitigate the human-machine conflict with the HMSD application.Furthermore,it could be further employed as an empirical trajectory prediction result.The algorithm employs the distribution state of the historical trajectory for human-like processing,simplifying the operational process and ensuring the credibility,integrity,and interpretability of the results.Moreover,in terms of optimization processing,the form of optimization search followed by collision avoidance detection is adopted to in principle reduce the calculation difficulty.Additionally,a new convex polygon collision detection method,namely the vertex embedding method,is proposed for collision avoidance detection.
