• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A route planning method comprising: selecting a path to be examined between the starting point and the destination point; - determining the vehicle independent environment conditions on the selected path; - determining a predictable rate of detection of landmarks for a location of the vehicle on the selected path, - determine a predictable location accuracy, and - determine whether predictable location accuracy is sufficient.
    公开号:FR3068777A1
    申请号:FR1856150
    申请日:2018-07-04
    公开日:2019-01-11
    发明作者:Holger Mielenz；Jan Rohde
    申请人:Robert Bosch GmbH；
    IPC主号:
    专利说明:

    The invention also relates to a vehicle equipped with an automated driving system implementing the route planning method.
    State of the art
    Current ADAS driving assistance systems (advanced driving assistance system) and highly automated driving systems (UAD systems) (automatic urban driving) require increasingly detailed knowledge of the vehicle environment and the perception of the situation. As a basis for the perception of the vehicle environment, we use sensor measurement data. From this data and using detection algorithms, objects can be extracted using which the environment of the vehicle is described and analyzed. Modern environmental sensors such as video cameras or laser scanners, in conjunction with detection algorithms, make it possible to capture a large amount of information about the environment of the vehicle, such as, for example, the shape of landmarks. These landmarks are traffic signals or signs, red lights, lane markings, etc. Detected objects or detected landmarks can be used to locate the vehicle. The available power of the entire automated driving system thus depends significantly on the capacity of the environmental sensors.
    Current assistance systems for guiding in a traffic lane depend on the reliability of detection and location of lane markings in relation to the vehicle. We know the document Borrmann, J.M. et al., STELLaR - a Case-Study on systematically embedding of traffic light récognition, Intelligent Transportation Systems (ITSC), 2014 IEEE, 17. International Conférence, pp. 1258, 1265, 8-11. 2014.
    This document describes a red light detection using object detection algorithms that are very demanding in terms of equipment.
    We also know the document Thrun, S., Finding
    Landmarks for Mobile Robot Navigation, Robotics & automation, 1998.
    Proceedings. 1998 IEEEE International Conférence, Tome 2, pp. 958
    963, 16.-20. May 1998.
    This document is a project to select landmarks to locate the vehicle.
    The accuracy of the vehicle location necessary for automated driving systems depends not only on the power of the environmental sensors, since the environmental conditions and the choice of the detection algorithm can have a significant influence. The power of the entire vehicle system is directly associated with the route to be taken and environmental conditions.
    Purpose of the invention
    The present invention aims to develop a route planning method for a motor vehicle equipped with an automated driving system and which takes into account the possibility of location specific to the route and the power of the automated driving system and thus allows planning a trip adapted to the situation.
    The invention also aims to develop a vehicle equipped with an automated driving system for route planning adapted to the situation.
    Presentation and advantages of the invention
    To this end, the subject of the invention is a route planning method for a vehicle equipped with an automated driving system, according to which an analysis of driving possibilities for at least one journey is established, this analysis of driving possibilities. comprising the following stages consisting in selecting a route to be examined between the starting point and the destination point, determining the environmental conditions independent of the vehicle on the chosen route, determining a predictable rate of detection of landmarks suitable for locating the vehicle on the selected route using environmental conditions independent of the vehicle, determine a predictable location accuracy on the selected route using the predictable detection rate and, determine if the predictable location accuracy is sufficient for driving assisted by the aut driving system on the selected route.
    The automated vehicle system is a driving assistance system to assist the driver and / or highly automated driving and / or autonomous driving.
    Advantageously, the route planning method takes account of a sufficiently precise location possibility with respect to the route and the available power of the system, in particular of the automated vehicle system. The routes along which, under the given environmental conditions and the environmental sensors available in the vehicle system, do not allow a location meeting demand, will be considered unusable. Such paths or path segments will then be bypassed. The method uses environmental conditions independent of the vehicle along the paths examined, these conditions constituting the input quantities. The marks for the location of the vehicle can be taken from a location map.
    The method makes it possible to plan journeys on which the power requested from the vehicle system is ensured with a high probability, which advantageously influences the robustness of the automated vehicle system. Compared to known route planning methods for automated vehicle systems, environment conditions independent of the vehicle are used here, i.e. environmental conditions which are not linked to the vehicle and / or the automated driving system. However, environmental conditions depending on the vehicle can also be included in the calculation of the predictable location accuracy. Vehicle-dependent environmental conditions are, for example, the efficiency of sensors or the robustness of algorithms for detecting landmarks.
    Environmental sensors can be video-stereo sensors or laser scanners or other suitable sensor devices.
    The step of determining the predictable detection rate of landmarks suitable for the location of the vehicle is not necessarily done as a separate process step, but it can also be implicitly included in the determination of the predictable location accuracy. The bottom line in the route planning process is the predictable location accuracy and the resulting determination of whether the predictable location accuracy for assisted driving in an automated vehicle system on the selected route is sufficient using the conditions depending on the vehicle on the selected route. For example, it is also possible that the determination of the predictable location accuracy or the determination of the location accuracy is sufficient to travel on the selected route with the assistance of the automated driving system, using a parametric model and / or neural network or other machine learning process and / or statistical models.
    In the case of a parameterized model, the detection rate can be found explicitly or implicit in the process. If the location accuracy is determined by a neural network or similar means, then the predictable detection rate is implicitly taken into account by the weighting learned by the neural network. In the case of a statistical model, the detection rate is taken into account by the weighted, statistical input, in determining the location accuracy. For the understanding of the invention it is important that the accuracy of the location depends implicitly or explicitly on the product of the detection rate, (that is to say on the percentage of landmark detected on the chosen path) and on the number of markers present on the chosen route.
    The detection rate implicitly or explicitly included in the method may also depend on the type of landmarks located in the selected path. For example, the rate of detection of traffic signals, traffic lane limitations, light signals, trees or construction may be variable. In addition, the detection rate may depend on the efficiency of the installation of surrounding field sensors. The detection algorithms used may have varying effectiveness for the detection rate of certain types of landmarks. All the factors mentioned above can intervene explicitly or implicitly in taking the detection rate into account, in determining the predictable location accuracy and the resulting determination as to whether the predictable location accuracy is sufficient. to travel the selected route with the assistance of an automated driving system.
    Preferably, the environmental conditions independent of the vehicle are the traffic density and / or the weather conditions and / or the road conditions.
    Weather conditions, traffic events and traffic density as well as road conditions can influence the detection characteristic of the sensors and thus influence the detection rate of different types of landmarks. In particular, a high traffic density can lead to a certain part of landmarks being hidden on the selected route, at least from time to time by vehicles preceding the vehicle equipped with the sensor; this will decrease the detection rate and thus the accuracy of the location. Advantageously, dependence on location accuracy and current weather conditions and traffic events can be taken into account. The determination of the environmental conditions, in particular the weather conditions and traffic events can be done, for example, by interrogating bands of meteorological data or bands of traffic data.
    Preferably, a rate of coverage of landmarks on the selected path is determined and this rate of coverage is preferably determined by using environmental conditions and / or the particular type of use of the environment and / or the appropriate road mark type; the accuracy of the location and / or the detection rate are determined using the coverage rate.
    Advantageously, the coverage rate influences the environmental conditions dependent on the journey and those independent of the journey on the accuracy of the location. The coverage rate may also depend on the type of road mark. Thus, for example, landmarks located at low height above the traffic lane are hidden from the environmental sensors because of traffic and vehicles, which consequently decreases the detection rate for such benchmarks. Faced with this, the detection rate for road markers installed at a significant height such as, for example, light signals is lower.
    The coverage rate can be taken into account implicitly or explicitly in the process. In the parametric models or the neural networks one can have an implicit taking into account by the parameters or the learned weights.
    Preferably, the coverage rate and / or the detection rate and / or the location accuracy are determined using a parameterized model; the parameterized model is preferably a machine learned model and further preferably the machine learned model has been established using previous driving possibility analyzes in particular using previous environmental conditions and / or previously determined detection and / or previously determined coverage rates and / or previously determined location accuracies and / or the power capacity of the sensor devices and / or detection algorithms of the automated vehicle system.
    Advantageously, the coverage rate and / or the detection rate and / or the accuracy of the location is determined using a parameterized model or a neural network. The parameterized model or the neural network thus determines the coverage rate and / or the detection rate and / or the accuracy of the localization based on analyzes of the possibility of driving made previously, that is to say using the results of analyzes of driving possibilities which were made before the current execution of the process.
    Correspondingly, it can be foreseen that before executing the analysis of driving possibilities, in particular before choosing a path to be examined, the parameterized model or the neural network performs a learning phase.
    According to a preferred embodiment, the coverage rate is determined using the parameterized model with, as input quantity of the parameterized model, the current meteorological information and / or the current traffic state on the selected route and / or the types of landmarks existing on the selected route.
    The system is based on a parameterized model with a machine learning process, in particular on a neural network which establishes the relationship between time and traffic data as well as other information concerning the environment, the perception modules of environment, used or environment sensors on the vehicle side and the resulting probability of detection of different types of landmarks.
    Preferably, the detection rate is determined using the coverage rate and / or a number, in particular a maximum number, and / or a digital density and / or the type of landmarks which can be detected on the selected route and / or the power of the sensor devices and / or the detection algorithms of the automated vehicle system and / or the environmental conditions and / or the accuracy of the location data, in particular of the GPS data.
    The coverage rate is used to determine the part of the landmark of a certain type that will probably be covered on the route selected by environmental conditions such as heavy traffic. The detection rate then depends on the coverage rate and also on the effectiveness of the sensors or sensor devices. In addition, the detection rate may depend on the effectiveness of the detection algorithms using different detection algorithms, especially for different types of landmarks.
    Preferably, the location accuracy is determined using a statistical model and this location accuracy is preferably determined using the predictable detection rate and / or the predictable coverage rate and / or the number and / or the digital density and / or the type of landmarks along the selected path.
    The location accuracy depends explicitly or implicitly on the foreseeable detection rate and the number or numerical density of the road markers according to the type of markers along the selected route. The detection rate can be influenced by the coverage rate.
    According to another advantageous characteristic, the determined location accuracy is compared with a predefined threshold and the driver of the vehicle will have the possibility of driving himself along the selected route and / or a new analysis of the possibilities of driving on another path if the location accuracy is below the threshold.
    While driving, if the location accuracy drops below the threshold, the driver will be informed that the selected route cannot be traveled by the vehicle with the automated vehicle system and that a new route is then planned.
    As a variant or in addition, the driver of the vehicle will have the possibility of deciding whether he wants to drive himself on the route initially selected and planned or whether he wishes to accept a possibly longer route.
    Advantageously, one can make a global planning and a change of planning of the whole journey for an automated vehicle system, towards a defined destination.
    Preferably, it is planned to determine if the predictable location accuracy is sufficient to travel on the selected route with an automated vehicle system, if the location possibility is sufficiently good and / or if this determination made on such a short route as possible is optimized with sufficient localization accuracy.
    In addition, preferably, the assisted driving with the automated vehicle system on the selected path determines suitable sensor devices and / or detector algorithms preferably if the location accuracy is greater than or equal to the threshold.
    Advantageously, the method determines which sensor device is suitable for which type of landmark, to travel on the selected path. Using a parameterized model, the system relies in particular on establishing the relationship between meteorological data, traffic data as well as other information relating to the environment, sensor devices or modules for perceiving the environment on the vehicle side and the resulting probability of detection of different types of landmarks. With the information concerning the density of landmarks which are, for example, extracted from a location map, and the types of landmarks likely to be detected and which are given by the location module of the vehicle driving system, the method makes it possible to determine whether the route can currently be taken by the vehicle system. For this, the method takes into account different elements of the vehicle-side detection algorithms, that is to say that, depending on the sensor device available or according to the detection algorithms available, certain types of vehicle may travel in areas in which others will not. It is thus possible to advantageously fix which sensor devices and / or which detection algorithms have been used to cover a path, in particular which segment of the path.
    The meteorological data, traffic information, information relating to the accuracy of the GPS system and information concerning the density of the landmarks detectable along the route are used as input quantities. A model whose parameters have been determined by a machine learning process will thus be used to obtain information concerning the selected route to determine if necessary with which detection algorithm the route can be carried out.
    Another solution to the problem of the invention is the production of a motor vehicle equipped with an automated driving system for implementing the method as described above.
    According to another advantageous characteristic, the automated vehicle system is designed for assisted driving and / or highly automated driving and / or autonomous driving.
    drawings
    The present invention will be described below, in more detail, using a route planning method for a motor vehicle, shown in the accompanying drawings in which:
    FIG. 1 shows a flowchart of a route planning method for a vehicle equipped with an automated driving system, FIG. 2 shows an overview of the input quantities for the learning phase of a model setting.
    Description of an embodiment of the invention
    Figure 1 shows a flowchart of a route planning method for a motor vehicle equipped with an automated driving system.
    In a first step SI optionally provided, during a learning phase of a parameterized model, in particular of a neural network, we rely on information such as meteorological data, detection rates, terrestrial landmarks of different types and other sources of information such as traffic data, a combination of these input quantities and the expected coverage rate for the different types of terrestrial landmark, preferably under given conditions. The position can, among other things, be determined by GPS, combined navigation or location on the vehicle side.
    In one case, the process begins with the analysis of the possibility of driving in step S2. Depending on the starting point and the destination, we select a route to be examined to take it with the automated vehicle system equipped with a navigation system.
    Then, in step S3, the environmental conditions independent of the vehicle on the selected path are determined. These environmental conditions can be current traffic events or current weather conditions on the selected route; they are obtained by interrogating a database. In addition, predictable landmarks are extracted from a location map.
    In the next step S4, the parameterized model is used to determine the expected coverage rate on the path under examination for each type of desired landmark. As an input quantity, current traffic events and / or current weather conditions are used, and where appropriate also the landmarks recorded in the location map for the selected route.
    In step S5, the rate of detection of landmarks for locating the vehicle on the selected path is determined using environmental conditions independent of the vehicle. For the determination of the detection rate, the coverage rate obtained in step S4 is used as input. In addition, account can be taken of vehicle-dependent conditions such as, for example, the efficiency of detection of different sensor devices such as stereo cameras or laser systems as well as the power of possible detection algorithms for different types of landmarks, especially in given environmental conditions (meteorology, traffic conditions,
    In step S6, apply the predictable location accuracy on the selected path using the predictable detection rate. In other words, with the combination of the determined coverage rate, the detection rate and the number, the numerical density and the type of landmarks along the route, the accuracy of the localization is evaluated using a statistical model.
    The accuracy of the location for the selected route results for a given type of landmarks, in particular from the product of the detection rate and the digital density of the landmarks existing on the selected route or of the segments of the selected route.
    In another step S7, it is determined whether the predictable location accuracy for driving with the assistance of the automated vehicle system is sufficient for the selected route. From the analysis of the possibility of driving, we decide on the need for a new route planning. If this was necessary, it can be done according to the process which begins with the new route selected by the analysis of the possibility of driving in step S2.
    If the accuracy of the location for driving assisted by the automated vehicle system of the selected route is sufficient, in another step S8, the method may indicate for a route considered to be usable, in addition to the sensor devices and the detection algorithms the vehicle system should use. In addition, a plan can be established to switch the detection algorithms along the selected path, to react to modified types of landmarks.
    For example, a route for an automated vehicle system is planned along a main road. Along this route, there is enough environmental information available so that the detection of all landmarks of the road sign type allows sufficiently precise location of the vehicle system.
    For traffic times on this route, it is nevertheless necessary to count with heavy traffic. An automated vehicle system implemented by the method determines this traffic condition and for the planned route, a detection rate which is lower than the foreseeable coverage rate which is based on environmental conditions independent of the vehicle and is significantly below 100%. However, the system determines as predictable location accuracy a value which is sufficiently good and is in particular above a predefined threshold. The system thus allows assisted driving by the automated vehicle system on the selected route.
    In another case, at the same time of day there can be a strong precipitation. The method again determines a detection rate. Due to bad weather conditions, this rate is not high enough, which results in particular from the increase in the coverage rate due to bad weather conditions. This does not allow for sufficient location accuracy. As part of the process, an alternative route can be selected and evaluated in the context of a new analysis of driving possibilities. This new route goes through less congested roads but with a lower coverage rate. The driver can be communicated as an option via a man-machine interface, that a longer trip has been planned and that there is only a possibility of selection between automated mode on this trip or non-automated driving on a shorter trip.
    FIG. 2 shows an overview of the input quantities for the learning phase of the parameterized model and in particular of the neural network.
    In the learning phase, the following are used as input parameters to the neural network 14:
    GPS 10 data,
    - traffic data 11,
    - meteorological data 12, and
    - previous detection rates 13.
    From the training phase, a model 15 is obtained for the probability depending on the type of detection of a landmark under different environmental conditions.
    权利要求:
    Claims (10)
    [1" id="c-fr-0001]
    1 °) Method for route planning for a vehicle equipped with an automated driving system, according to which an analysis of driving possibilities is established for at least one journey, this analysis of driving possibilities comprising the following steps consisting in:
    select a route to be examined between the starting point and the destination point, determine the environmental conditions independent of the vehicle on the chosen route, determine a predictable rate of detection of landmarks suitable for locating the vehicle on the selected route in using environmental conditions independent of the vehicle, determining a predictable location accuracy on the selected route using the predictable detection rate, and determining whether the predictable location accuracy is sufficient for assisted driving by the automated driving system on the selected route.
    [2" id="c-fr-0002]
    2 °) Method according to claim 1, characterized in that the environmental conditions independent of the vehicle are the density of traffic and / or weather conditions and / or road conditions on the chosen route.
    [3" id="c-fr-0003]
    3 °) Method according to claims 1 or 2, characterized in that one determines the coverage rate of landmarks on the selected route, * the coverage rate being determined preferably using the environmental conditions and / or the type appropriate landmark, * the accuracy of the location and / or the detection rate being determined using the coverage rate.
    [4" id="c-fr-0004]
    4 °) Method according to claim 3, characterized in that the coverage rate and / or the detection rate and / or the location accuracy are determined using a parameterized model,
    [5" id="c-fr-0005]
    5 * the parameterized model preferably being a machine learned model, moreover, the preferably machine learned model is learned using analyzes of possibility of prior driving, in particular using previous environmental conditions and / or previously determined detection rates and / or previously determined coverage rates and / or previously determined location accuracies and / or the available power of the sensor devices and / or detection algorithms of the automated driving system.
    5 °) Method according to one of the preceding claims, characterized in that the detection rate is determined by using the coverage rate 20 and / or a number in particular maximum and / or a density of numbers and / or types of detectable landmarks on the selected route and / or the possible power of the sensor devices and / or algorithms for detecting the automated driving system and / or environmental conditions and / or the accuracy of the location data, in particular the data GPS.
    [6" id="c-fr-0006]
    6 °) Method according to one of the preceding claims, characterized in that the location accuracy is determined using a statistical model, * the location accuracy being preferably obtained using the detection rate predictable and / or predictable coverage rate and / or the number and / or numerical density and / or type of landmarks along the selected route.
    [7" id="c-fr-0007]
    7 °) Method according to one of the preceding claims, characterized in that the location accuracy determined is compared to a predefined threshold, * the driver of the vehicle is allowed to drive manually on the selected route and / or a new analysis of possibilities for driving another route if the accuracy of the location is less than that of the first threshold.
    [8" id="c-fr-0008]
    8 °) Method according to one of the preceding claims, characterized in that the appropriate sensor devices for the assisted driving of the automated driving system on the selected path and / or the detector algorithms are determined, preferably if the accuracy of the location is greater than or equal to the threshold.
    [9" id="c-fr-0009]
    9 °) Motor vehicle equipped with an automated driving system allowing the implementation of a route planning method according to any one of claims 1 to 8.
    [10" id="c-fr-0010]
    10 °) Motor vehicle according to claim 9, in which the driving assistance system is produced for assisted driving and / or highly automated driving and / or autonomous driving.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3068777A1|2019-01-11|METHOD FOR PLANNING THE TRACK OF A MOTOR VEHICLE EQUIPPED WITH AN AUTOMATED DRIVING SYSTEM AND VEHICLE IMPLEMENTING THE METHOD
    US10471955B2|2019-11-12|Stop sign and traffic light alert
    DK2619059T3|2017-02-06|Driver profiling system and method
    EP2079984B1|2014-06-11|Method of constructing a digital road model
    CN101964941A|2011-02-02|Intelligent navigation and position service system and method based on dynamic information
    FR2902909A1|2007-12-28|METHOD FOR DETERMINING THE RUNNING LIMITS OF A VEHICLE
    CN107636751A|2018-01-26|Traveling lane discriminating gear and traveling lane method of discrimination
    US20170074670A1|2017-03-16|Travel time data processing apparatus, travel time data processing method and program
    US20190360820A1|2019-11-28|Method and device for executing at least one measure for increasing the safety of a vehicle
    RU2017127535A|2019-02-05|SYSTEM FOR OVERCASTING THE ICE CAPACITY PLACES ON THE UPCOMING BRIDGE AND THE RELATED METHOD
    SE539156C2|2017-04-18|Map structure in a vehicle
    EP3472015B1|2020-06-10|Method for determining a reference driving class
    US20210088340A1|2021-03-25|Inferring lane boundaries via high speed vehicle telemetry
    US10876843B2|2020-12-29|Method, device and system for wrong-way driver detection
    JP4521695B2|2010-08-11|Link travel time estimation apparatus, program and method
    US10532750B2|2020-01-14|Method, device and system for wrong-way driver detection
    EP3792127A1|2021-03-17|Method for assisting the choice of positioning of a vehicle on a road comprising a plurality of lanes
    US20210048819A1|2021-02-18|Apparatus and method for determining junction
    CN113847925A|2021-12-28|Method, device, equipment and medium for detecting vehicle yaw based on track data
    FR2956639A1|2011-08-26|Device for estimating driving of vehicle e.g. lorry, to provide driving energy for vehicle during long journey, has driving center providing energy to vehicle while measuring variation between actual parameter and reference parameter
    JP2020204531A|2020-12-24|Map information generator, map information generation method, and map information generation program
    FR3080448A1|2019-10-25|DEVICE AND METHOD FOR ANALYZING THE POSITION OF A VEHICLE BY COMPARISON OF DETERMINED AND KNOWN ENVIRONMENTAL INFORMATION
    Akai et al.2013|Mitigating location and speed errors in floating car data using context-based accuracy estimation
    FR3089187A1|2020-06-05|Driving aid method and device for a motor vehicle
    FR3103305A1|2021-05-21|Method and device for predicting at least one dynamic characteristic of a vehicle at a point in a road segment.
    同族专利:
    公开号 | 公开日
    CN109211255A|2019-01-15|
    DE102017211556A1|2019-01-10|
    FR3068777B1|2021-09-17|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    PL2490092T3|2011-02-16|2014-02-28|Siemens Ag|Method for autonomous localisation of a driver-less motorised vehicle|
    DE102013205840A1|2013-04-03|2014-10-09|Robert Bosch Gmbh|Automatic driving on a route|
    EP2915718B1|2014-03-04|2018-07-11|Volvo Car Corporation|Apparatus and method for continuously establishing a boundary for autonomous driving availability and an automotive vehicle comprising such an apparatus|
    JP6537780B2|2014-04-09|2019-07-03|日立オートモティブシステムズ株式会社|Traveling control device, in-vehicle display device, and traveling control system|
    EP3109115B1|2015-06-23|2018-01-31|Volvo Car Corporation|Arrangement and method for facilitating handover to and from an automated autonomous driving aid system|
    CN106908775B|2017-03-08|2019-10-18|同济大学|A kind of unmanned vehicle real-time location method based on laser reflection intensity|CN109808615A|2019-03-22|2019-05-28|北京经纬恒润科技有限公司|A kind of automatic Pilot method and device|
    WO2021016920A1|2019-07-31|2021-02-04|深圳市大疆创新科技有限公司|Method, system and device for identifying accessibility, and computer-readable storage medium|
    CN111024084A|2019-12-17|2020-04-17|国汽智能网联汽车研究院有限公司|Automatic driving method, device, equipment and storage medium for automatic driving vehicle|
    法律状态:
    2019-07-24| PLFP| Fee payment|Year of fee payment: 2 |
    2020-07-27| PLFP| Fee payment|Year of fee payment: 3 |
    2020-11-06| PLSC| Search report ready|Effective date: 20201106 |
    2021-07-22| PLFP| Fee payment|Year of fee payment: 4 |
    优先权:
    申请号 | 申请日 | 专利标题
    DE102017211556.4A|DE102017211556A1|2017-07-06|2017-07-06|Method for route planning for a motor vehicle with an automated vehicle system and motor vehicle with an automated vehicle system|
    DE102017211556.4|2017-07-06|
    [返回顶部]