技术领域Technical field
本发明属于智能驾驶中的场景识别和数据挖掘领域，具体涉及了一种跟车加速场景数据采集、挖掘及特征分析方法和系统。The invention belongs to the field of scene recognition and data mining in intelligent driving, and specifically relates to a method and system for data collection, mining and feature analysis of car-following acceleration scenes.
背景技术Background technique
车辆跟车行为主要描述单车道上车辆列队行驶的条件下，随道路交通流密度变化，自车跟随所在车道内前车行驶且受前车影响的驾驶行为。车辆跟车加速行为是指自车与前车处于稳定跟车状态的前提下，前车突然加速驶离致使两车的相对距离增加而引起的后车驾驶员的加速反应，是驾驶员典型驾驶行为之一，也是跟车启停，距离调控等智能驾驶功能的重要数据支撑。针对这一驾驶行为展开深入的研究有助于相关智能驾驶辅助技术的开发。Vehicle following behavior mainly describes the driving behavior in which vehicles follow the vehicle in front of them in the lane and are affected by the vehicle in front of them as the density of road traffic flow changes under the condition of platooning of vehicles on a single lane. Vehicle following acceleration behavior refers to the acceleration reaction of the driver behind the vehicle caused by the sudden acceleration of the vehicle in front and the increase in the relative distance between the two vehicles when the vehicle in front and the vehicle in front are in a stable following state. It is a typical driving behavior of the driver. One of the behaviors is also an important data support for intelligent driving functions such as car start and stop, distance control, etc. In-depth research on this driving behavior will help the development of related intelligent driving assistance technologies.
当前智能驾驶已经进入数据驱动的时代，能够通过高效率、低成本获取海量驾驶场景库、数据库是驱动智能驾驶技术迭代的基本。目前自然驾驶场景库中关于跟车加速场景的数据并不丰富，且存在场景边界不清晰，场景误提取及漏提取率较高，场景识别提取效率低，场景信息不全面，相关特征分析内容不足等问题，难以表征驾驶员驾驶行为特征，不足以支撑智能驾驶决策规系统的深度学习和优化，无法满足现有智能驾驶技术的功能安全性、类人性、舒适性等相关技术开发的需求。Currently, intelligent driving has entered a data-driven era. Being able to obtain massive driving scene libraries and databases with high efficiency and low cost is the basis for driving the iteration of intelligent driving technology. At present, the natural driving scene library does not have abundant data on car-following acceleration scenarios, and there are unclear scene boundaries, high scene mis-extraction and missing extraction rates, low scene recognition and extraction efficiency, incomplete scene information, and insufficient relevant feature analysis content. It is difficult to characterize the driver's driving behavior characteristics, is insufficient to support the deep learning and optimization of intelligent driving decision-making systems, and cannot meet the needs of functional safety, human-likeness, comfort and other related technology development needs of existing intelligent driving technology.
发明内容Contents of the invention
本发明的目的是提供一种跟车加速场景数据采集、挖掘及特征分析方法和系统，以解决跟车加速场景的数据集不丰富，场景识别提取效率低，场景边界不清晰，场景误提取及漏提取率较高，场景信息不全面，相关特征分析内容不足的问题。The purpose of the present invention is to provide a method and system for data collection, mining and feature analysis of car-following acceleration scenarios to solve the problem of insufficient data sets for car-following acceleration scenarios, low efficiency of scene recognition and extraction, unclear scene boundaries, incorrect scene extraction and The missing extraction rate is high, the scene information is incomplete, and the relevant feature analysis content is insufficient.
为实现本发明的目的，本发明提供的技术方案如下：In order to achieve the purpose of the present invention, the technical solutions provided by the present invention are as follows:
第一方面first
本发明提供了一种跟车加速场景数据采集、挖掘及特征分析方法，包括以下步骤：The invention provides a car-following acceleration scene data collection, mining and feature analysis method, which includes the following steps:
步骤S1：采集自车行车信息及环境车辆状态信息；Step S1: Collect self-vehicle driving information and environmental vehicle status information;
步骤S2：对自车行车信息及环境车辆状态信息进行数据清洗处理，获得清洗处理后的数据；Step S2: Perform data cleaning processing on the self-vehicle driving information and environmental vehicle status information, and obtain the cleaned data;
步骤S3：根据清洗处理后的数据，判断自车驾驶行为是否满足跟车约束条件，并对满足跟车约束条件的跟车场景进行标记；Step S3: Based on the cleaned data, determine whether the self-driving behavior satisfies the car-following constraints, and mark the car-following scenes that satisfy the car-following constraints;
步骤S4：对标记的跟车场景施加跟车加速约束条件，判断标记的跟车场景是否满足跟车加速约束条件，若满足，则从当前标记的跟车场景中，提取满足要求的跟车加速场景片段；Step S4: Apply the car following acceleration constraint to the marked car following scene, and determine whether the marked car following scene satisfies the car following acceleration constraint. If so, extract the car following acceleration that meets the requirements from the currently marked car following scene. scene clips;
步骤S5：提取所有满足要求的跟车加速场景片段的特征参数，汇总并生成特征参数表格，基于特征参数表格，通过特征参数统计图表查看特征参数分布情况，根据所述特征参数分布情况获取跟车加速场景的驾驶行为特征。Step S5: Extract the characteristic parameters of all car-following acceleration scene clips that meet the requirements, summarize and generate a feature parameter table, check the feature parameter distribution through the feature parameter statistical chart based on the feature parameter table, and obtain the car-following based on the feature parameter distribution. Driving behavior characteristics in acceleration scenarios.
第二方面Second aspect
与上述方法相对应地，本发明还提供了一种跟车加速场景数据采集、挖掘及特征分析系统，包括信息采集单元、数据清洗处理单元以及跟车场景标记单元、跟车加速场景片段提取单元、驾驶行为特征获取单元；Corresponding to the above method, the present invention also provides a car-following acceleration scene data collection, mining and feature analysis system, including an information collection unit, a data cleaning processing unit, a car-following scene marking unit, and a car-following acceleration scene segment extraction unit. , Driving behavior characteristic acquisition unit;
所述信息采集单元用于采集自车行车信息及环境车辆状态信息；The information collection unit is used to collect self-vehicle driving information and environmental vehicle status information;
所述数据清洗处理单元用于对自车行车信息及环境车辆状态信息进行数据清洗处理，获得清洗处理后的数据；The data cleaning processing unit is used to perform data cleaning processing on the vehicle driving information and environmental vehicle status information, and obtain the cleaned data;
所述跟车场景标记单元用于根据清洗处理后的数据，判断自车驾驶行为是否满足跟车约束条件，并对满足跟车约束条件的跟车场景进行标记；The car-following scene marking unit is used to determine whether the driving behavior of the self-vehicle satisfies the car-following constraint conditions based on the cleaned data, and to mark the car-following scenes that satisfy the car-following constraint conditions;
所述跟车加速场景片段提取单元用于对标记的跟车场景施加跟车加速约束条件，判断标记的跟车场景是否满足跟车加速约束条件，若满足，则从当前标记的跟车场景中，提取满足要求的跟车加速场景片段；The car-following acceleration scene segment extraction unit is used to apply car-following acceleration constraints to the marked car-following scenes, and determine whether the marked car-following scenes satisfy the car-following acceleration constraints. , extract car-following acceleration scene clips that meet the requirements;
所述驾驶行为特征获取单元用于提取所有满足要求的跟车加速场景片段的特征参数，汇总并生成特征参数表格，基于特征参数表格，通过特征参数统计图表查看特征参数分布情况，根据所述特征参数分布情况获取跟车加速场景的驾驶行为特征。The driving behavior feature acquisition unit is used to extract the feature parameters of all car-following acceleration scene segments that meet the requirements, summarize and generate a feature parameter table, and based on the feature parameter table, check the feature parameter distribution through the feature parameter statistical chart. According to the features The parameter distribution is used to obtain the driving behavior characteristics of car-following acceleration scenarios.
与现有技术相比，本发明具有的技术优势如下：Compared with the existing technology, the technical advantages of the present invention are as follows:
（1）本发明提供的方案能够自车及场景数据采集及存储，可以保证数据的有效性及丰富性；(1) The solution provided by the present invention can collect and store vehicle and scene data, and can ensure the validity and richness of the data;
（2）本发明提供的方案能够实现数据清洗及自动提取，相比人工数据清洗及场景提取，可极大提高跟车加速场景的识别精度及提取效率，通过大幅降低人力消耗，还可降低场景提取的成本；(2) The solution provided by the present invention can realize data cleaning and automatic extraction. Compared with manual data cleaning and scene extraction, it can greatly improve the recognition accuracy and extraction efficiency of car-following acceleration scenes. By greatly reducing manpower consumption, it can also reduce scene complexity. cost of extraction;
（3）本发明提供的方案能够实现特征参数表格的及特征参数统计图表的生成，能够直观的、快速的获取跟车加速场景的驾驶行为特征，对智能驾驶技术的开发及优化具有重要意义。(3) The solution provided by the present invention can realize the generation of characteristic parameter tables and characteristic parameter statistical charts, and can intuitively and quickly obtain the driving behavior characteristics of car-following acceleration scenarios, which is of great significance to the development and optimization of intelligent driving technology.
附图说明Description of the drawings
图1为本发明实施例提供的方法流程示意图；Figure 1 is a schematic flow chart of a method provided by an embodiment of the present invention;
图2为本发明实施例提供的信息采集单元布置示意图；Figure 2 is a schematic layout diagram of an information collection unit provided by an embodiment of the present invention;
图3为本发明实施例提供的弯道工况下车辆位置示意图；Figure 3 is a schematic diagram of the vehicle position under curve conditions provided by the embodiment of the present invention;
图4为本发明实施例提供的跟车加速场景示意图；Figure 4 is a schematic diagram of a car-following acceleration scenario provided by an embodiment of the present invention;
图5为本发明实施例所述的特征参数散点图示意图；Figure 5 is a schematic diagram of a scatter plot of characteristic parameters according to the embodiment of the present invention;
图中，1-毫米波雷达；2-显示器；3-车辆CAN总线；4-功能摄像头；5-数据存储器。In the figure, 1-millimeter wave radar; 2-display; 3-vehicle CAN bus; 4-functional camera; 5-data memory.
具体实施方式Detailed ways
下面结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.
如图1所示，本发明提供了一种跟车加速场景数据采集、挖掘及特征分析方法，包括以下步骤：As shown in Figure 1, the present invention provides a car-following acceleration scene data collection, mining and feature analysis method, which includes the following steps:
步骤S1：采集自车行车信息及环境车辆状态信息；Step S1: Collect self-vehicle driving information and environmental vehicle status information;
其中，所述自车行车信息包括自车车长、自车车宽以及自车速度、自车加速度、方向盘转角、距离车道线的距离；Wherein, the self-vehicle driving information includes the self-vehicle length, the self-vehicle width, the self-vehicle speed, the self-vehicle acceleration, the steering wheel angle, and the distance from the lane line;
所述环境车辆状态信息包括环境车辆ID、环境车辆类型、环境车辆车长、环境车辆车宽以及环境车辆与自车相对距离、环境车辆与自车相对角度、环境车辆与自车相对速度。The environmental vehicle status information includes environmental vehicle ID, environmental vehicle type, environmental vehicle length, environmental vehicle width, relative distance between the environmental vehicle and the own vehicle, relative angle between the environmental vehicle and the own vehicle, and relative speed between the environmental vehicle and the own vehicle.
步骤S2：对自车行车信息及环境车辆状态信息进行数据清洗处理，获得清洗处理后的数据；其中，所述数据清洗处理包括数据填充与滤波处理；Step S2: Perform data cleaning processing on the self-vehicle driving information and environmental vehicle status information to obtain cleaned data; wherein the data cleaning processing includes data filling and filtering processing;
所述数据填充具体为：对于自车行车信息及环境车辆状态信息中Na值连续且数量达到预设高阈值的片段，默认为系统未识别并未采集到信息且将Na值替换为固定float型数值；对于自车行车信息及环境车辆状态信息中Na值未达到预设低阈值的片段，默认为数据异常短暂丢失，将Na值通过线性插值的方法进行填充替换。The data filling is specifically as follows: for segments with continuous Na values in the bicycle driving information and environmental vehicle status information and the number reaches the preset high threshold, the default is that the system has not recognized and collected the information and replaces the Na value with a fixed float type. Numerical value; for segments where the Na value in the self-vehicle driving information and environmental vehicle status information does not reach the preset low threshold, the default is that the data is abnormal and temporarily lost, and the Na value is filled and replaced by linear interpolation.
所述滤波处理具体为：采用对称指数移动平均法去除噪声。The filtering process specifically includes: using a symmetric exponential moving average method to remove noise.
由于自车行车信息及环境车辆状态信息中存在一定的噪声，通过滤波的方法去除噪声，使数据更平滑、可用性更强。Since there is a certain amount of noise in the vehicle driving information and environmental vehicle status information, filtering is used to remove the noise to make the data smoother and more usable.
步骤S3：根据清洗处理后的数据，判断自车驾驶行为是否满足跟车约束条件，并对满足跟车约束条件的跟车场景进行标记；Step S3: Based on the cleaned data, determine whether the self-driving behavior satisfies the car-following constraints, and mark the car-following scenes that satisfy the car-following constraints;
其中，所述判断自车驾驶行为是否满足跟车约束条件，具体如下：Among them, the determination of whether the self-driving behavior satisfies the following constraints is as follows:
步骤S3.1：判断是否存在跟车目标车，具体如下：Step S3.1: Determine whether there is a following target car, the details are as follows:
如图3所示为本发明实施例提供的弯道工况下车辆位置示意图。Figure 3 shows a schematic diagram of the vehicle position under curve conditions according to an embodiment of the present invention.
根据环境车辆与自车相对距离、环境车辆与自车相对角度/>，计算环境车辆与自车相对横向距离/>、环境车辆与自车相对纵向距离/>，计算方法如下：The relative distance between the vehicle and its own vehicle according to the environment , the relative angle between the environment vehicle and the self-vehicle/> , calculate the relative lateral distance between the environment vehicle and the own vehicle/> , the relative longitudinal distance between the environment vehicle and the self-vehicle/> , the calculation method is as follows:
； ;
对弯道情况下环境车辆与自车相对横向距离进行修正，弯道曲率半径为r，遵循左正右负准则，修正后的环境车辆与自车相对横向距离为/>，修正方法如下：The relative lateral distance between the environment vehicle and the self-vehicle under curve conditions Correction is made, the radius of curvature of the curve isr, following the left positive and right negative criterion, the corrected relative lateral distance between the environment vehicle and the self-vehicle is/> , the correction method is as follows:
； ;
根据环境车辆与自车相对纵向距离和修正后的环境车辆与自车相对横向距离，确定处于自车道内且位于自车前方的环境车辆为跟车目标车；According to the relative longitudinal distance between the environmental vehicle and its own vehicle and the corrected relative lateral distance between the environment vehicle and the self-vehicle , determine that the environmental vehicle in the own lane and in front of the own vehicle is the following target vehicle;
步骤S3.2：若存在跟车目标车，则判断是否满足跟车约束条件，跟车约束条件如下：Step S3.2: If there is a following target vehicle, determine whether the following constraints are met. The following constraints are as follows:
； ;
其中，为自车与环境车辆的纵向距离最大距离约束值，/>为跟车目标车的宽度，/>为车道线的宽度，/>为自车前轴中心距离右侧车道线中心的长度，/>为自车前轴中心距离左侧车道线中心的长度，/>为跟车场景辅助系数，若跟车场景满足跟车约束条件，/>输出值为1，否则输出值为0。in, is the maximum distance constraint value of the longitudinal distance between the self-vehicle and the surrounding vehicles,/> is the width of the following target car,/> is the width of the lane line,/> is the length from the center of the front axle of your vehicle to the center of the right lane line,/> is the length from the center of the front axle of your vehicle to the center of the left lane line,/> is the auxiliary coefficient of the car following scene. If the car following scene satisfies the car following constraints,/> The output value is 1, otherwise the output value is 0.
步骤S4：对标记的跟车场景施加跟车加速约束条件，判断标记的跟车场景是否满足跟车加速约束条件，若满足，则从当前标记的跟车场景中，提取满足要求的跟车加速场景片段；Step S4: Apply the car following acceleration constraint to the marked car following scene, and determine whether the marked car following scene satisfies the car following acceleration constraint. If so, extract the car following acceleration that meets the requirements from the currently marked car following scene. scene clips;
其中，所述判断标记的跟车场景是否满足跟车加速约束条件，具体如下：Among them, the determination of whether the marked car-following scene satisfies the car-following acceleration constraint is as follows:
步骤S4.1：在满足跟车约束条件的前提下，确定自车加速时机t1，准则如下：Step S4.1: Under the premise that the vehicle following constraints are met, determine the self-vehicle acceleration timing t1. The criteria are as follows:
； ;
其中，为自车加速度，/>为自车加速度率，所述自车加速度率通过对自车加速度求导获取；/>为自车加速度预设阈值；当且仅当自车加速度大于自车加速度预设阈值/>且自车加速度率/>为正时，确定为自车加速时机t1；in, is the acceleration of the own vehicle,/> is the self-vehicle acceleration rate, which is obtained by derivation of the self-vehicle acceleration;/> Preset the threshold for the own vehicle's acceleration; if and only if the own vehicle's acceleration Greater than the self-vehicle acceleration preset threshold/> And the own vehicle acceleration/> When it is positive, it is determined as the own vehicle acceleration timing t1;
步骤S4.2：从自车加速时机t1起对自车加速度进行周期循环判断，当判断第一个自车加速度大于0的时刻，确定为自车加速开始时刻t2；从自车加速时机t1对自车加速度进行周期循环判断，当判断自车第一个加速度小于0的时刻，确定为跟车加速场景结束时刻t3；Step S4.2: Make periodic judgments on the self-vehicle acceleration starting from the self-vehicle acceleration timing t1. When the first self-vehicle acceleration is judged to be greater than 0, it is determined as the self-vehicle acceleration start time t2; starting from the self-vehicle acceleration timing t1, The acceleration of the own vehicle is judged periodically. When the first acceleration of the own vehicle is judged to be less than 0, it is determined as the end time t3 of the vehicle-following acceleration scene;
步骤S4.3：确定跟车加速场景开始时刻t4，准则如下：Step S4.3: Determine the starting time t4 of the car-following acceleration scene. The criteria are as follows:
； ;
其中，为跟车目标车加速度，/>为跟车目标车加速度预设阈值，为跟车目标车加速度率；其中，从自车加速场景开始时刻t2对目标车加速进行周期循环判断，当判断第一个跟车目标车加速度大于跟车目标车加速度预设阈值且跟车目标车加速度率为正时，确定为跟车加速场景开始时刻t4；in, is the acceleration of the following target vehicle,/> Preset threshold value for the acceleration of the following target vehicle, is the acceleration rate of the following target vehicle; among them, the target vehicle acceleration is judged periodically from the starting time t2 of the self-vehicle acceleration scene. When it is judged that the acceleration of the first following target vehicle is greater than the preset threshold value of the following target vehicle acceleration and the following target vehicle When the vehicle acceleration rate is positive, it is determined as the starting time t4 of the car-following acceleration scene;
步骤S4.4：判断跟车加速场景开始时刻t4到跟车加速场景结束时刻t3之间的时间间隔是否满足预设的跟车场景提取范围阈值，若满足则判定当前标记的跟车场景满足跟车加速约束条件；若不满足，则判定当前标记的跟车场景不满足跟车加速约束条件，从跟车加速场景结束时刻t3继续开始，继续进行下一个标记的跟车场景的判断，直到判断完所有标记的跟车场景。Step S4.4: Determine whether the time interval between the start time t4 of the car-following acceleration scene and the end time t3 of the car-following acceleration scene meets the preset car-following scene extraction range threshold. If it does, determine whether the currently marked car-following scene satisfies the following car-following scene. The car acceleration constraint conditions; if not satisfied, it is determined that the currently marked car following scene does not meet the car following acceleration constraint conditions. Starting from the end time t3 of the car following acceleration scene, the judgment of the next marked car following scene continues until the judgment is made. Complete all marked car following scenarios.
其中，跟车目标车加速度计算如下，获取目标车辆相对速度，采样时长，根据线性回归的方法对目标车辆的加速情况进行估计，计算公式如下：Among them, the acceleration of the following target vehicle The calculation is as follows. Obtain the relative speed of the target vehicle, the sampling duration, and estimate the acceleration of the target vehicle based on the linear regression method. The calculation formula is as follows:
； ;
； ;
对于已有的跟车目标车速度信息的时间序列进行线性回归，为跟车目标车速度估计值，/>为截距，n为数据序列的长度，回归方程中/>代表跟车目标车速度信息的时间序列，回归方程中/>代表跟车目标车速度信息序列，/>为时间序列的平均值，/>为速度信息的平均值，斜率/>代表了跟车目标车加速度/>的情况。Perform linear regression on the time series of the existing speed information of the following target vehicle, is the estimated speed of the following target vehicle,/> is the intercept, n is the length of the data sequence, in the regression equation/> Represents the time series of speed information of the following target vehicle, in the regression equation/> Represents the following target vehicle speed information sequence,/> is the average of the time series,/> is the average value of speed information, slope/> Represents the acceleration of the following target vehicle/> Case.
步骤S5：提取所有满足要求的跟车加速场景片段的特征参数，汇总并生成特征参数表格，基于特征参数表格，通过特征参数统计图表查看特征参数分布情况，根据所述特征参数分布情况获取跟车加速场景的驾驶行为特征。Step S5: Extract the characteristic parameters of all car-following acceleration scene clips that meet the requirements, summarize and generate a feature parameter table, check the feature parameter distribution through the feature parameter statistical chart based on the feature parameter table, and obtain the car-following based on the feature parameter distribution. Driving behavior characteristics in acceleration scenarios.
其中，所述特征参数统计图为散点图、直方图和/或箱线图等。Wherein, the characteristic parameter statistical graph is a scatter plot, a histogram and/or a box plot, etc.
如图5所示，取400例跟车加速场景片段中加速时机的关键参数生成散点图，如自车速度ego_v(km/h)、自车距离跟车目标车的纵向距离obj_d(m)，对散点进行拟合得到ego_v(km/h)与obj_d(m)的关系式如下：As shown in Figure 5, the key parameters of the acceleration timing in 400 car-following acceleration scene clips are taken to generate a scatter plot, such as the self-vehicle speed ego_v (km/h), the self-vehicle distance and the longitudinal distance of the following target car obj_d (m) , and the relationship between ego_v(km/h) and obj_d(m) is obtained by fitting the scatter points as follows:
； ;
其中，x表示自车速度，y表示自车距离跟车目标车纵向距离，R²表示拟合系数。基本可以反映大多数驾驶员跟车加速时机的自车速度与自车距离跟车目标车纵向距离的关系。可以实现对智能驾驶技术的功能安全性、类人性、舒适性等相关技术优化及开发的需求。Among them, x represents the speed of the own vehicle, y represents the longitudinal distance between the own vehicle and the target vehicle, and R² represents the fitting coefficient. It can basically reflect the relationship between the self-vehicle speed and the distance between the self-vehicle and the longitudinal distance of the target vehicle when most drivers follow the car. It can realize the needs for optimization and development of related technologies such as functional safety, human-likeness, and comfort of intelligent driving technology.
本发明还提供了一种跟车加速场景数据采集、挖掘及特征分析系统，包括信息采集单元、数据清洗处理单元以及跟车场景标记单元、跟车加速场景片段提取单元、驾驶行为特征获取单元；The invention also provides a car-following acceleration scene data collection, mining and feature analysis system, which includes an information collection unit, a data cleaning processing unit, a car-following scene marking unit, a car-following acceleration scene segment extraction unit, and a driving behavior feature acquisition unit;
所述信息采集单元用于采集自车行车信息及环境车辆状态信息；The information collection unit is used to collect self-vehicle driving information and environmental vehicle status information;
所述数据清洗处理单元用于对自车行车信息及环境车辆状态信息进行数据清洗处理，获得清洗处理后的数据；The data cleaning processing unit is used to perform data cleaning processing on the vehicle driving information and environmental vehicle status information, and obtain the cleaned data;
所述跟车场景标记单元用于根据清洗处理后的数据，判断自车驾驶行为是否满足跟车约束条件，并对满足跟车约束条件的跟车场景进行标记；The car-following scene marking unit is used to determine whether the driving behavior of the self-vehicle satisfies the car-following constraint conditions based on the cleaned data, and to mark the car-following scenes that satisfy the car-following constraint conditions;
所述跟车加速场景片段提取单元用于对标记的跟车场景施加跟车加速约束条件，判断标记的跟车场景是否满足跟车加速约束条件，若满足，则从当前标记的跟车场景中，提取满足要求的跟车加速场景片段；The car-following acceleration scene segment extraction unit is used to apply car-following acceleration constraints to the marked car-following scenes, and determine whether the marked car-following scenes satisfy the car-following acceleration constraints. , extract car-following acceleration scene clips that meet the requirements;
所述驾驶行为特征获取单元用于提取所有满足要求的跟车加速场景片段的特征参数，汇总并生成特征参数表格，基于特征参数表格，通过特征参数统计图表查看特征参数分布情况，根据所述特征参数分布情况获取跟车加速场景的驾驶行为特征。The driving behavior feature acquisition unit is used to extract the feature parameters of all car-following acceleration scene segments that meet the requirements, summarize and generate a feature parameter table, and based on the feature parameter table, check the feature parameter distribution through the feature parameter statistical chart. According to the features The parameter distribution is used to obtain the driving behavior characteristics of car-following acceleration scenarios.
需要说明的是，如图2所示，所述信息采集单元包括传感器、采集车以及数据存储器等，传感器包括毫米波雷达1和功能摄像头4，分别固定于采集车上并与数据存储器5相连，同时数据存储器5还与车辆CAN总线3以及显示器2连接，进而可以识别、采集、存储并显示自车及环境车辆的驾驶行为信息，包括自车速度、自车加速度、目标车类型、目标车速度、目标车相对距离等相关信息。It should be noted that, as shown in Figure 2, the information collection unit includes a sensor, a collection vehicle, a data memory, etc. The sensor includes a millimeter wave radar 1 and a functional camera 4, which are respectively fixed on the collection vehicle and connected to the data memory 5. At the same time, the data memory 5 is also connected to the vehicle CAN bus 3 and the display 2, thereby identifying, collecting, storing and displaying the driving behavior information of the own vehicle and surrounding vehicles, including the own vehicle speed, the own vehicle acceleration, the target vehicle type, and the target vehicle speed. , relative distance of the target vehicle and other related information.
其中，所述判断自车驾驶行为是否满足跟车约束条件，具体如下：Among them, the determination of whether the self-driving behavior satisfies the following constraints is as follows:
步骤S3.1：判断是否存在跟车目标车，具体如下：Step S3.1: Determine whether there is a following target car, the details are as follows:
根据环境车辆与自车相对距离、环境车辆与自车相对角度/>，计算环境车辆与自车相对横向距离/>、环境车辆与自车相对纵向距离/>，计算方法如下：The relative distance between the vehicle and its own vehicle according to the environment , the relative angle between the environment vehicle and the self-vehicle/> , calculate the relative lateral distance between the environment vehicle and the own vehicle/> , the relative longitudinal distance between the environment vehicle and the self-vehicle/> , the calculation method is as follows:
； ;
对弯道情况下环境车辆与自车相对横向距离进行修正，弯道曲率半径为r，遵循左正右负准则，修正后的环境车辆与自车相对横向距离为/>，修正方法如下：The relative lateral distance between the environment vehicle and the self-vehicle under curve conditions Correction is made, the radius of curvature of the curve isr, following the left positive and right negative criterion, the corrected relative lateral distance between the environment vehicle and the self-vehicle is/> , the correction method is as follows:
； ;
根据环境车辆与自车相对纵向距离和修正后的环境车辆与自车相对横向距离，确定处于自车道内且位于自车前方的环境车辆为跟车目标车；According to the relative longitudinal distance between the environmental vehicle and its own vehicle and the corrected relative lateral distance between the environment vehicle and the self-vehicle , determine that the environmental vehicle in the own lane and in front of the own vehicle is the following target vehicle;
步骤S3.2：若存在跟车目标车，则判断是否满足跟车约束条件，跟车约束条件如下：Step S3.2: If there is a following target vehicle, determine whether the following constraints are met. The following constraints are as follows:
； ;
其中，为自车与环境车辆的纵向距离最大距离约束值，/>为跟车目标车的宽度，/>为车道线的宽度，/>为自车前轴中心距离右侧车道线中心的长度，/>为自车前轴中心距离左侧车道线中心的长度，/>为跟车场景辅助系数，若跟车场景满足跟车约束条件，/>输出值为1，否则输出值为0。in, is the maximum distance constraint value of the longitudinal distance between the self-vehicle and the surrounding vehicles,/> is the width of the following target car,/> is the width of the lane line,/> is the length from the center of the front axle of your vehicle to the center of the right lane line,/> is the length from the center of the front axle of your vehicle to the center of the left lane line,/> is the auxiliary coefficient of the car following scene. If the car following scene satisfies the car following constraints,/> The output value is 1, otherwise the output value is 0.
其中，所述判断标记的跟车场景是否满足跟车加速约束条件，具体如下：Among them, the determination of whether the marked car-following scene satisfies the car-following acceleration constraint is as follows:
步骤S4.1：在满足跟车约束条件的前提下，确定自车加速时机t1，准则如下：Step S4.1: Under the premise that the vehicle following constraints are met, determine the self-vehicle acceleration timing t1. The criteria are as follows:
； ;
其中，其中，为自车加速度，/>为自车加速度率，所述自车加速度率通过对自车加速度求导获取；/>为自车加速度预设阈值；当且仅当自车加速度大于自车加速度预设阈值/>且自车加速度率/>为正时，确定为自车加速时机t1；Among them, among them, is the acceleration of the own vehicle,/> is the self-vehicle acceleration rate, which is obtained by derivation of the self-vehicle acceleration;/> Set a threshold for the own vehicle’s acceleration; if and only if the own vehicle’s acceleration Greater than the self-vehicle acceleration preset threshold/> And the own vehicle acceleration/> When it is positive, it is determined as the own vehicle acceleration timing t1;
步骤S4.2：从自车加速时机t1起对自车加速度进行周期循环判断，当判断第一个自车加速度大于0的时刻，确定为自车加速开始时刻t2；从自车加速时机t1对自车加速度进行周期循环判断，当判断自车第一个加速度小于0的时刻，确定为跟车加速场景结束时刻t3；Step S4.2: Make periodic judgments on the self-vehicle acceleration starting from the self-vehicle acceleration timing t1. When the first self-vehicle acceleration is judged to be greater than 0, it is determined as the self-vehicle acceleration start time t2; starting from the self-vehicle acceleration timing t1, The acceleration of the own vehicle is judged periodically. When the first acceleration of the own vehicle is judged to be less than 0, it is determined as the end time t3 of the vehicle-following acceleration scene;
步骤S4.3：确定跟车加速场景开始时刻t4，准则如下：Step S4.3: Determine the starting time t4 of the car-following acceleration scene. The criteria are as follows:
； ;
其中，为跟车目标车加速度，/>为跟车目标车加速度预设阈值，为跟车目标车加速度率；其中，从自车加速场景开始时刻t2对目标车加速进行周期循环判断，当判断第一个跟车目标车加速度大于跟车目标车加速度预设阈值且跟车目标车加速度率为正时，确定为跟车加速场景开始时刻t4；in, is the acceleration of the following target vehicle,/> Preset threshold value for the acceleration of the following target vehicle, is the acceleration rate of the following target vehicle; among them, the target vehicle acceleration is judged periodically from the starting time t2 of the self-vehicle acceleration scene. When it is judged that the acceleration of the first following target vehicle is greater than the preset threshold value of the following target vehicle acceleration and the following target vehicle When the vehicle acceleration rate is positive, it is determined as the starting time t4 of the vehicle following acceleration scene;
步骤S4.4：判断跟车加速场景开始时刻t4到跟车加速场景结束时刻t3之间的时间间隔是否满足预设的跟车场景提取范围阈值，若满足则判定当前标记的跟车场景满足跟车加速约束条件；若不满足，则判定当前标记的跟车场景不满足跟车加速约束条件，从跟车加速场景结束时刻t3继续开始，继续进行下一个标记的跟车场景的判断，直到判断完所有标记的跟车场景。Step S4.4: Determine whether the time interval between the start time t4 of the car-following acceleration scene and the end time t3 of the car-following acceleration scene meets the preset car-following scene extraction range threshold. If it does, determine whether the currently marked car-following scene satisfies the following car-following scene. The car acceleration constraint conditions; if not satisfied, it is determined that the currently marked car following scene does not meet the car following acceleration constraint conditions. Starting from the end time t3 of the car following acceleration scene, the judgment of the next marked car following scene continues until the judgment is made. Complete all marked car following scenarios.
最后应当说明的是：上述实施例只是用于对本发明的举例和说明，而非意在将本发明限制于所描述的实施例范围内。此外本领域技术人员可以理解的是，本发明不局限于上述实施例，根据本发明教导还可以做出更多种的变型和修改，这些变型和修改均落在本发明所要求保护的范围内。Finally, it should be noted that the above-mentioned embodiments are only used to illustrate and illustrate the present invention, and are not intended to limit the present invention to the scope of the described embodiments. In addition, those skilled in the art can understand that the present invention is not limited to the above embodiments, and more variations and modifications can be made according to the teachings of the present invention, and these variations and modifications all fall within the scope of protection claimed by the present invention. .
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311587866.2ACN117312776B (en) | 2023-11-27 | 2023-11-27 | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311587866.2ACN117312776B (en) | 2023-11-27 | 2023-11-27 | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data |
| Publication Number | Publication Date |
|---|---|
| CN117312776A CN117312776A (en) | 2023-12-29 |
| CN117312776Btrue CN117312776B (en) | 2024-03-12 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311587866.2AActiveCN117312776B (en) | 2023-11-27 | 2023-11-27 | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data |
| Country | Link |
|---|---|
| CN (1) | CN117312776B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103754224A (en)* | 2014-01-24 | 2014-04-30 | 清华大学 | Vehicle multi-target coordinating lane changing assisting adaptive cruise control method |
| CN109816811A (en)* | 2018-10-31 | 2019-05-28 | 杭州云动智能汽车技术有限公司 | A kind of nature driving data acquisition device |
| CN111710176A (en)* | 2020-06-08 | 2020-09-25 | 同济大学 | An intersection signal-vehicle trajectory cooperative control method in a vehicle-road cooperative environment |
| CN112466119A (en)* | 2020-11-26 | 2021-03-09 | 清华大学 | Method and system for predicting vehicle following speed of vehicle by using vehicle-road cooperative data |
| CN113487874A (en)* | 2021-05-27 | 2021-10-08 | 中汽研（天津）汽车工程研究院有限公司 | System and method for collecting, identifying and classifying following behavior scene data |
| CN114492043A (en)* | 2022-01-27 | 2022-05-13 | 吉林大学 | A Personalized Driver-Following Modeling Approach Considering Perceptually Restricted Properties |
| CN114973733A (en)* | 2022-04-29 | 2022-08-30 | 北京交通大学 | Method for optimizing and controlling track of networked automatic vehicle under mixed flow at signal intersection |
| CN115366914A (en)* | 2022-08-03 | 2022-11-22 | 清华大学苏州汽车研究院（相城） | Method, apparatus, and medium for controlling autonomous vehicle based on model predictive control |
| CN115713863A (en)* | 2022-10-31 | 2023-02-24 | 长沙理工大学 | Trajectory planning and signal optimization method for special phase for automatic driving |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103754224A (en)* | 2014-01-24 | 2014-04-30 | 清华大学 | Vehicle multi-target coordinating lane changing assisting adaptive cruise control method |
| CN109816811A (en)* | 2018-10-31 | 2019-05-28 | 杭州云动智能汽车技术有限公司 | A kind of nature driving data acquisition device |
| CN111710176A (en)* | 2020-06-08 | 2020-09-25 | 同济大学 | An intersection signal-vehicle trajectory cooperative control method in a vehicle-road cooperative environment |
| CN112466119A (en)* | 2020-11-26 | 2021-03-09 | 清华大学 | Method and system for predicting vehicle following speed of vehicle by using vehicle-road cooperative data |
| CN113487874A (en)* | 2021-05-27 | 2021-10-08 | 中汽研（天津）汽车工程研究院有限公司 | System and method for collecting, identifying and classifying following behavior scene data |
| CN114492043A (en)* | 2022-01-27 | 2022-05-13 | 吉林大学 | A Personalized Driver-Following Modeling Approach Considering Perceptually Restricted Properties |
| CN114973733A (en)* | 2022-04-29 | 2022-08-30 | 北京交通大学 | Method for optimizing and controlling track of networked automatic vehicle under mixed flow at signal intersection |
| CN115366914A (en)* | 2022-08-03 | 2022-11-22 | 清华大学苏州汽车研究院（相城） | Method, apparatus, and medium for controlling autonomous vehicle based on model predictive control |
| CN115713863A (en)* | 2022-10-31 | 2023-02-24 | 长沙理工大学 | Trajectory planning and signal optimization method for special phase for automatic driving |
| Title |
|---|
| "基于深度强化学习的驾驶员跟车模型研究";郭景华 等;《汽车工程》;参见正文第1-2章* |
| Publication number | Publication date |
|---|---|
| CN117312776A (en) | 2023-12-29 |
| Publication | Publication Date | Title |
|---|---|---|
| CN112487908B (en) | Detection and dynamic tracking method of front vehicle line-pressing behavior based on vehicle video | |
| CN111324120A (en) | A method for automatic driving front car cut-in and cut-out scene extraction | |
| CN111599181A (en) | Typical natural driving scene recognition and extraction method for intelligent driving system test | |
| CN112950811B (en) | New energy automobile region operation risk assessment and early warning system integrating whole automobile safety | |
| CN109767619B (en) | An intelligent network-connected pure electric vehicle driving condition prediction method | |
| CN111724603B (en) | CAV state determination method, device, device and medium based on traffic trajectory data | |
| CN113195323B (en) | Safe autonomous driving with detected objects | |
| CN112706785B (en) | Method and device for selecting cognitive target of driving environment of automatic driving vehicle and storage medium | |
| CN110458050B (en) | Vehicle cut-in detection method and device based on vehicle-mounted video | |
| CN112373474B (en) | Lane line fusion and transverse control method, system, vehicle and storage medium | |
| CN113306558B (en) | Lane changing decision method and system based on lane changing interaction intention | |
| CN103136341A (en) | Lane line reconstruction device based on Bezier curve | |
| WO2023197408A1 (en) | Method and apparatus for determining vehicle speed control model training sample | |
| CN115635961A (en) | Sample data generation method and trajectory prediction method and device applying same | |
| CN114643993A (en) | Lane merging point calculation system, lane merging point calculation method, vehicle and storage medium | |
| CN118025155A (en) | A method, device, vehicle and program product for determining a vehicle-following target | |
| CN114301792B (en) | Traffic flow simulation method and traffic flow sensor | |
| US10906553B2 (en) | Systems and methods for vehicle acceleration event prediction inhibit | |
| CN117312776B (en) | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data | |
| CN112598314B (en) | Method, device, equipment and medium for determining perception confidence of intelligent driving automobile | |
| CN115431981B (en) | Driving auxiliary identification system based on high-precision map | |
| CN117246333A (en) | A vehicle driving demand prediction method based on near-field predictive information | |
| CN112948525B (en) | High-precision dynamic map generation method, equipment and medium for intelligent networked automobile | |
| CN114999150A (en) | Road congestion determination method, device, vehicle and storage medium | |
| CN116469263A (en) | Traffic flow control method considering bus stop under network environment |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |