摘要：【中文】本发明公开了一种基于合流时间优化的匝道合流协同控制方法及系统，该方法包括：步骤1，按照各个智能网联车辆驶入匝道合流区域的先后顺序进行编号；步骤2，当智能网联车辆驶入匝道合流区域时，计算抵达匝道口合流处的参考合流时间、最早合流时间和实际合流时间；步骤3，每辆智能网联车辆利用无线通信将自身的身份编号、位置信息及行驶至匝道口合流处的实际合流时间向其它智能网联车辆进行广播；步骤4，构造每辆智能网联车辆实际合流时间的优化问题；步骤5，更新自身的实际合流时间，直至收敛；步骤6，即控制自车速度于收敛所得的实际合流时间通过匝道口合流处。本发明可实现对智能网联车辆合流时间的优化，提升匝道合流的安全性和通行效率。 【EN】The invention discloses a ramp confluence cooperative control method and system based on confluence time optimization, wherein the method comprises the following steps: step 1, numbering according to the sequence of driving-in ramp confluence areas of all intelligent network-connected vehicles; step 2, when the intelligent network-connected vehicle drives into the ramp confluence area, calculating the reference confluence time, the earliest confluence time and the actual confluence time of the confluence position of the arrival ramp port; step 3, each intelligent networked vehicle broadcasts the identity number and the position information of the intelligent networked vehicle and the actual merging time of the intelligent networked vehicle driving to the merging position of the ramp port to other intelligent networked vehicles by utilizing wireless communication; step 4, constructing an optimization problem of the actual confluence time of each intelligent networked vehicle; step 5, updating the actual confluence time of the self until convergence; and 6, controlling the speed of the vehicle to the actual merging time obtained by convergence to pass through the merging position of the ramp port. The method can optimize the converging time of the intelligent networked vehicles, and improve the safety and the passing efficiency of the ramp converging.

摘要：【中文】本发明公开了一种融合车载信息的自动驾驶系统激光雷达在线标定方法，其目的是：提供一种端到端的激光雷达外参在线标定方法，避免复杂的数学模型推导和优化，避免使用额外设备，利用深度卷积神经网络的数据分析能力处理激光雷达的点云数据和车辆ECU数据，在线实时估计出激光雷达的外参误差，从而实现对激光雷达外参的实时修正，提升自动驾驶系统环境感知功能的准确性和稳定性，保证自动驾驶系统的行车安全。 【EN】The invention discloses an automatic driving system laser radar online calibration method fusing vehicle-mounted information, which aims to: the method for calibrating the laser radar external parameters from end to end on line is provided, complex mathematical model derivation and optimization are avoided, extra equipment is avoided, point cloud data and vehicle ECU (electronic control unit) data of the laser radar are processed by using the data analysis capability of a deep convolution neural network, and external parameter errors of the laser radar are estimated on line in real time, so that real-time correction of the laser radar external parameters is realized, the accuracy and the stability of the environment sensing function of an automatic driving system are improved, and the driving safety of the automatic driving system is ensured.

摘要：【中文】本发明的一种基于深度卷积神经网络的自动驾驶系统激光雷达在线标定方法，其目的是：提供一种端到端的激光雷达外参在线标定方法，避免复杂的数学模型推导和优化，利用深度卷积神经网络的数据分析能力处理激光雷达的点云数据和车辆GNSS数据，在线实时估计出激光雷达的外参误差，从而实现对激光雷达外参的实时修正，提升自动驾驶系统环境感知功能的准确性和稳定性，保证自动驾驶系统的行车安全。 【EN】The invention discloses an automatic driving system laser radar online calibration method based on a deep convolutional neural network, which aims to: the method for calibrating the laser radar external parameters from end to end on line is provided, complex mathematical model derivation and optimization are avoided, the point cloud data and vehicle GNSS data of the laser radar are processed by utilizing the data analysis capability of the deep convolution neural network, and the external parameter errors of the laser radar are estimated on line in real time, so that the real-time correction of the laser radar external parameters is realized, the accuracy and the stability of the environment sensing function of the automatic driving system are improved, and the driving safety of the automatic driving system is ensured.

摘要：【中文】本发明公开了一种基于IMU预积分的自动驾驶系统激光雷达外参估计方法，其目的是：基于自动驾驶系统原有车载传感器，提供一种端到端的激光雷达外参在线标定方法，避免复杂的数学模型推导和优化，利用深度卷积神经网络的数据分析能力处理激光雷达的点云数据和车辆状态轨迹数据，在线实时估计出激光雷达的外参误差，从而实现对激光雷达外参的实时修正，提升自动驾驶系统环境感知功能的准确性和稳定性，保证自动驾驶系统的行车安全。 【EN】The invention discloses an automatic driving system laser radar external parameter estimation method based on IMU pre-integration, which aims to: based on the original vehicle-mounted sensor of the automatic driving system, an end-to-end laser radar external parameter online calibration method is provided, complex mathematical model derivation and optimization are avoided, point cloud data and vehicle state track data of the laser radar are processed by using the data analysis capacity of a deep convolution neural network, and external parameter errors of the laser radar are estimated online in real time, so that real-time correction of the laser radar external parameters is achieved, accuracy and stability of an environment sensing function of the automatic driving system are improved, and driving safety of the automatic driving system is guaranteed.

摘要：【中文】本发明公开了一种车边云协同架构的协同任务迁移系统及算法，包括：车载终端集群，该车载终端集群用于采集车辆信息和输出车辆信息至边缘服务器及远程云服务器；边缘服务器，设置于路侧边缘，与车载终端集群通信连接，以接收并处理车载终端集群采集处理的车辆信息和输出信号指令至车载终端集群内；远程云服务器，与车载终端集群通信连接，与车载终端集群之间进行身份认证、数据存储和数据收发。本发明的车边云协同架构的协同任务迁移系统，通过车载终端、边缘计算服务器和远程云服务器的组合作用，便可实现从传统的车‑边协同架构、车‑云协同架构到车边云协同架构的演进，选择最优的任务迁移机制，实现车载的能耗最小，可有效延长车辆的续航里程。 【EN】The invention discloses a cooperative task migration system and algorithm of a car-side cloud cooperative architecture, which comprises the following steps: the vehicle-mounted terminal cluster is used for acquiring vehicle information and outputting the vehicle information to the edge server and the remote cloud server; the edge server is arranged at the edge of the road side and is in communication connection with the vehicle-mounted terminal cluster so as to receive and process vehicle information acquired and processed by the vehicle-mounted terminal cluster and output a signal instruction to the vehicle-mounted terminal cluster; and the remote cloud server is in communication connection with the vehicle-mounted terminal cluster and performs identity authentication, data storage and data transceiving with the vehicle-mounted terminal cluster. According to the cooperative task migration system of the vehicle-side cloud cooperative architecture, the evolution from the traditional vehicle-side cooperative architecture, the vehicle-cloud cooperative architecture to the vehicle-side cloud cooperative architecture can be realized through the combined action of the vehicle-mounted terminal, the edge computing server and the remote cloud server, the optimal task migration mechanism is selected, the vehicle-mounted energy consumption is minimum, and the cruising mileage of the vehicle can be effectively prolonged.

摘要：【中文】本发明公开了一种车辆路径跟踪控制方法，该方法包括：S1，根据已有的参考路径点，获得一条路径点更密集的新参考路径；S2，获得车辆状态信息；S3，在新参考路径上找出最近路径点；S4，以最近路径点为起点，在新参考路径上向车辆行驶的前方搜索N个预瞄点；S5，构建预测模型、目标函数以及系统约束，根据当前测量信息和预测模型，预测车辆未来动态，在线求解满足所述目标函数和约束条件的优化问题，获取N个预瞄点所对应的期望前轮转向角构成的最优控制序列；S6，根据最优控制序列，控制车辆直到下一采样时刻到达，下一观测时刻到达时，重复步骤S2至S5。本发明提供的方法跟踪精度较高，同时也能够保证在控制过程中的舒适性，不会产生控制量的突变。 【EN】The invention discloses a vehicle path tracking control method, which comprises the following steps: s1, obtaining a new reference path with denser path points according to the existing reference path points; s2, obtaining vehicle state information; s3, finding out the nearest path point on the new reference path; s4, searching N preview points in front of the vehicle on the new reference path by taking the nearest path point as a starting point; s5, constructing a prediction model, an objective function and system constraints, predicting future dynamics of the vehicle according to current measurement information and the prediction model, solving an optimization problem meeting the objective function and constraint conditions on line, and acquiring an optimal control sequence formed by expected front wheel steering angles corresponding to N pre-aiming points; and S6, controlling the vehicle until the next sampling time arrives according to the optimal control sequence, and repeating the steps S2 to S5 when the next observation time arrives. The method provided by the invention has higher tracking precision, can ensure the comfort in the control process and cannot generate sudden change of the control quantity.

摘要：【中文】本发明公开了一种保障交通流稳定的车辆编队控制方法，包括如下步骤：步骤1，将编队中车辆从前到后进行1～N编号，其中第1辆为领航车辆，其余N‑1辆为跟随车辆，步骤2，在行驶过程中，领航车辆和每个跟随车辆i利用车载通信将自身的编号、位置、速度、加速度信息发送至后方至多m个车辆；步骤3，每个跟随车辆i利用自身和所接收的前方至多m个车辆的位置、速度、加速度信息计算期望加速度；步骤4，每个跟随车辆进行底层加速与制动执行器控制；步骤5，每个跟随车辆不断重复上述步骤2～4。本发明的保障交通流稳定的车辆编队控制方法，可通过车辆之间的通信，在干扰存在下保持车辆编队期望几何构型。 【EN】The invention discloses a vehicle formation control method for guaranteeing traffic flow stability, which comprises the following steps: step 1, numbering vehicles in a formation from front to back by 1-N, wherein the 1 st vehicle is a pilot vehicle, and the rest N-1 vehicles are following vehicles, and step 2, in the driving process, the pilot vehicle and each following vehicle i send the number, position, speed and acceleration information of the pilot vehicle and each following vehicle to at most m vehicles behind by using vehicle-mounted communication; step 3, each following vehicle i calculates expected acceleration by utilizing the position, speed and acceleration information of the following vehicle i and the received at most m vehicles in front; step 4, each following vehicle performs bottom layer acceleration and brake actuator control; and 5, continuously repeating the steps 2-4 for each following vehicle. The vehicle formation control method for guaranteeing the stability of the traffic flow can keep the expected geometric configuration of the vehicle formation in the presence of interference through communication between vehicles.

摘要：【中文】本发明公开了一种基于路灯杆的智慧交通控制方法及系统，该方法包括：当检测到的路灯杆L_n在t时刻检测到的交通流信息λ_nt小于交通流阈值λ时，开启边缘控制模式：判断路灯杆L_n在t时刻检测到的动态目标数量D_nt是否大于0，如果是，则向路灯杆L_n及其前方路灯杆中的边缘计算单元发送提高灯光亮度的控制指令，如果不是，则判断路灯杆L_n在t时刻检测到的动态目标数量D_nt＝0持续的时间是否达到预设时间，如果已达到，则向该路灯杆L_n发送降低灯光亮度的控制指令。本发明能够根据交通的实际情形，灵活控制照明模块的灯光使用，这样有利于节约电能，降低交通成本，符合绿色环保的要求。 【EN】The invention discloses an intelligent traffic control method and system based on a light pole, wherein the method comprises the following steps: when detected light pole L_nTraffic flow information λ detected at time t_ntAnd when the value is smaller than the traffic flow threshold lambda, starting an edge control mode: judge light pole L_nNumber of dynamic objects D detected at time t_ntWhether is greater than 0, if yes, then to light pole L_nAnd the edge calculation unit in the street lamp pole in front of the edge calculation unit sends a control instruction for improving the light brightness, and if not, the street lamp pole L is judged_nNumber of dynamic objects D detected at time t_ntWhether the duration time reaches the preset time or not is 0, and if the duration time reaches the preset time, the street lamp pole is turned onL_nAnd sending a control instruction for reducing the brightness of the lamp light. The invention can flexibly control the light of the lighting module to be used according to the actual situation of traffic, thus being beneficial to saving electric energy, reducing traffic cost and meeting the requirement of environmental protection.

摘要：【中文】本发明公开了一种基于图搜索和几何曲线融合的路径规划方法，包括如下步骤：步骤1：得到地图信息，确定车辆的起始点和目标点；步骤2：通过Hybrid A*算法对节点进行拓展，得到新的节点及其状态信息；步骤3：判断新节点在直行状态下，是否能与目标点所在射线相交；步骤4：判断新节点能否通过几何曲线路径到达目标点；步骤5：判断ProState集合是否为空集。本发明的基于图搜索和几何曲线融合的路径规划方法，解决了传统的Hybrid A*无法精确到达目标点、无法满足车辆目标横摆角要求的问题。 【EN】The invention discloses a path planning method based on graph search and geometric curve fusion, which comprises the following steps: step 1: obtaining map information, and determining a starting point and a target point of a vehicle; step 2: expanding the nodes through a Hybrid A-algorithm to obtain new nodes and state information thereof; and step 3: judging whether the new node can intersect the ray where the target point is located in the straight-going state; and 4, step 4: judging whether the new node can reach a target point through the geometric curve path; and 5: and judging whether the ProState set is an empty set. The path planning method based on graph search and geometric curve fusion solves the problems that the traditional Hybrid A can not accurately reach a target point and can not meet the requirement of a vehicle target yaw angle.

摘要：【中文】本发明公开了一种露天矿区交叉路口车辆通行控制方法，该方法能够为露天矿区车辆通行交叉路口时提供安全有效的通行控制方法，将能够根据车辆的装载状态与道路坡度情况，确定不同的预约距离，从而给予车辆不同的通行交叉路口优先权，有利于减少安全隐患和提高露天矿区通行效率。 【EN】The invention discloses a method for controlling vehicle passing at an intersection of an open-pit mine area, which can provide a safe and effective passing control method for vehicles in the open-pit mine area to pass through the intersection, and can determine different reserved distances according to the loading state of the vehicles and the road gradient condition, thereby giving the vehicles different passing intersections priority, being beneficial to reducing potential safety hazards and improving the passing efficiency of the open-pit mine area.