WO2019210599A1 - 一种停车位识别方法及泊车方法 - Google Patents
一种停车位识别方法及泊车方法 Download PDFInfo
- Publication number
- WO2019210599A1 WO2019210599A1 PCT/CN2018/099191 CN2018099191W WO2019210599A1 WO 2019210599 A1 WO2019210599 A1 WO 2019210599A1 CN 2018099191 W CN2018099191 W CN 2018099191W WO 2019210599 A1 WO2019210599 A1 WO 2019210599A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- parking space
- parking
- space information
- vehicle
- information
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000006073 displacement reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/06—Automatic manoeuvring for parking
Definitions
- the invention relates to the field of automatic parking, in particular to a parking space recognition method, a parking method and a vehicle navigation.
- Parking space identification is an important sensing unit of the automatic parking system. At present, most vehicles with automatic parking function adopt pure ultrasonic radar parking space recognition or panoramic parking space recognition.
- the pure ultrasonic radar parking space recognition scheme detects the obstacle information around the vehicle through the long-distance radar, and simultaneously processes the detected data and outputs the parking space information.
- the scheme relies heavily on obstacles around the vehicle, and the effect of the final parking of the vehicle is referenced by the surrounding obstacles, which makes it difficult to meet the requirements of standard parking using automatic parking in many cases.
- the panoramic parking space recognition scheme also known as the AVM parking space recognition scheme, performs image recognition processing on the parking line of the ground through the panoramic system, identifies the target parking spaces that can be parked, and outputs corresponding parking space information.
- the scheme can achieve more accurate parking effects, but for parking spaces with no special parking spaces or parking spaces, the scheme will not be able to identify parking spaces.
- the present invention provides a parking space recognition method and a parking method.
- a parking space identification method includes the following steps:
- S20 Perform an ultrasonic parking space recognition algorithm to identify and draw a parking space, and obtain first parking space information
- S30 Send the first parking space information to the automatic parking system, and execute the AVM parking space recognition algorithm to continuously identify the parking space when the parking is performed. If the parking space information is recognized during the parking, the parking space information is used as The second parking space information is sent to the automatic parking system instead of the first parking space information.
- step 20 includes the following sub-steps:
- step S23 Determine whether the distance between the coordinate point B1 and the coordinate point C1 meets the parking standard. If yes, execute the first parking space information generating step to generate the first parking space information, otherwise return to step S21.
- the first parking space information includes the following sub-steps:
- step S20 the enrolling step is also performed simultaneously:
- the AVM parking space recognition algorithm is executed. If the parking space information is recognized, the parking space information is used as the second parking space information, and the second parking space information is transmitted to the automatic parking system; the first parking space information is ignored.
- first parking space information and or the second parking space information includes current vehicle position coordinates and parking space edge coordinates.
- the present invention also provides a parking method comprising the following steps:
- A20 Send the higher parking space information of the priority that can be obtained to the automatic parking system
- A30 Continuously execute the first parking space recognition algorithm and the second parking space recognition algorithm, and execute an automatic parking step, during which the higher priority parking space information is sent to when the higher priority parking space information is acquired.
- Automatic parking system for automatic parking
- step A40 Detect whether the automatic parking step is finished, or return to step A30.
- the first parking space recognition algorithm is an ultrasonic parking space recognition algorithm
- the second parking space algorithm is an AVM parking space recognition algorithm
- step A30 the vehicle speed and the user command are also detected.
- the step A30 is performed, otherwise the waiting for detecting step is performed.
- the waiting for detecting step includes the following substeps:
- step B20 Detect whether the current vehicle position exceeds the effective parking range. If not, return to step B10, and if yes, prompt the user and return to step A10.
- step A10 cannot search for the valid parking space information
- the prompt for the parking space search identification is returned.
- the accuracy of the ultrasonic parking space recognition problem can be effectively solved, and the parking standardity of the automatic parking can be improved, and the AVM can be compensated for the parking spaces with no special parking spaces or parking spaces. Defects.
- FIG. 1 is a schematic diagram of a parking space recognition method according to Embodiment 1 of the present invention.
- FIG. 2 is a schematic diagram showing the principle of an ultrasonic parking space recognition algorithm according to Embodiment 1 of the present invention.
- FIG. 3 is a schematic diagram of a parking space recognition method according to Embodiment 2 of the present invention.
- FIG. 4 is a schematic diagram of an automatic parking method according to Embodiment 3 of the present invention.
- a parking space recognition method combines AVM parking space recognition and ultrasonic radar parking space recognition to more effectively and accurately identify the parking space, and solves the inaccuracy of the prior art summary ultrasonic radar parking space recognition, and the AVM parking space. Identify problems with small scope and identification limitations. Specifically, as shown in FIG. 1 , the following steps are included:
- the track algorithm of the embodiment specifically measures the wheel displacement by using the wheel speed pulse information returned by the wheel speed sensor, according to the change of the vehicle displacement during the sampling period.
- the vehicle azimuth increment and the position increment are calculated to obtain the pose of the vehicle relative to the initial state, and after continuous operation, this can obtain the track of the vehicle.
- S20 Perform an ultrasonic parking space recognition algorithm to identify and draw a parking space, and obtain first parking space information. Specifically, as shown in FIG. 2, it includes the following sub-steps:
- the car drives forward.
- the obstacle inspection vehicle is carried out around the vehicle along the driving direction of the vehicle.
- the user can select the operation to determine the key monitoring direction, such as selecting the parking space on the left side of the vehicle.
- the form of the coordinates may be two-dimensional coordinates or angular coordinates, and usually two-dimensional coordinates are used.
- the algorithm continues to collect the obstacle coordinate point.
- the ultrasonic radar will lose the detection of the obstacle, and the obstacle is lost.
- the coordinate point collected at the moment of object detection is called the coordinate point B1 of the first lost obstacle.
- the obstacle is continuously detected, and when the obstacle is searched for the second time, the coordinates of the point are recorded and set as the coordinate point C1. Referring to Figure 2, from a general perspective, the position between the coordinate point B1 and the coordinate point C1 is likely to be a parking space.
- Step S23 In order to further determine whether the parking space meets the requirements and whether the width is sufficient for parking, it may be determined whether the distance between the coordinate point B1 and the coordinate point C1 is greater than a certain ratio of the vehicle width, and if yes, the first parking space information is generated. Steps: Generate first parking space information, otherwise return to step S21.
- the generating the first parking space information includes the following sub-steps:
- the current coordinates of the vehicle are set as the origin by the conversion of the coordinates, and the coordinates of the coordinate point B1 and the coordinate point C1 are converted into the coordinate point B2 and the coordinate point C2 with respect to the current position of the vehicle by the coordinate difference.
- the center point of the rear axle of the automobile can be used as the calculation origin to ensure more accurate parking.
- the AVM parking space recognition algorithm continuously identifies the parking space.
- the recognition range can be set near the range of the coordinate point B1 and the coordinate point C1, thereby making the AVM recognition more purposeful.
- the parking space information is used as the second parking space information with higher priority.
- the second parking space information is automatically sent to the automatic parking system instead of the first parking space information, and the automatic parking system adopts a higher priority.
- the second parking space information of the class is used as the parking basis.
- the format of the first parking space information is the same, and the second parking space information also includes current vehicle position coordinates and parking space edge coordinates.
- the difference between the embodiment and the embodiment 1 is that, as shown in FIG. 3, in order to improve the efficiency of the automatic parking and the corresponding speed, the embodiment also performs the step S20 at the same time.
- the AVM parking space recognition algorithm will be executed. And at the same time, the two kinds of parking space information acquisition conditions are detected.
- the AVM parking space recognition algorithm directly recognizes the parking space information as the second parking space information and the second parking, regardless of whether the ultrasonic parking space recognition algorithm recognizes the parking space information.
- the bit information is sent to the automatic parking system, ignoring the first parking space information. It can be understood that step S30 can be directly discarded.
- the ultrasonic parking space recognition algorithm can also be continuously executed to continuously acquire the first parking space information.
- the embodiment provides a parking method based on the first embodiment and the second embodiment, and includes the following steps:
- the first parking space recognition algorithm can make the recognition range wider parking space recognition method for preliminary determination of the parking space, and the second parking space recognition algorithm can cooperate with the method with higher recognition accuracy.
- the parking space recognition algorithm with a wide recognition range will first recognize the parking space, thereby improving the recognition speed of the second parking space recognition algorithm.
- the first parking space recognition algorithm may be, but is not limited to, an ultrasonic parking space recognition algorithm
- the second parking space algorithm may be limited to the AVM parking space recognition algorithm.
- the technical solutions disclosed in this embodiment will be described later by using the above two algorithms, as shown in FIG. 4 .
- A20 In order to improve the accuracy of the automatic parking, it is necessary to transmit the higher parking space information of the priority that can be obtained to the automatic parking system. That is, when the first parking space information and the second parking space information are simultaneously acquired, the second parking space information with higher priority is transmitted to the automatic parking system, and the system uses the second parking space information as the position determination basis. .
- the first parking space information is used when the second parking space information cannot be recognized.
- the vehicle parking search recognition prompt will be returned, and at the same time, whether to return to search for the parking space according to the user's operation can be selected.
- the automatic parking step is performed according to the received parking space information, and at the same time, the first parking space recognition algorithm and the second parking space recognition algorithm are continuously executed, during which time
- the parking space information of the priority position is higher than the currently used parking space information
- the parking space information of the higher priority is sent to the automatic parking system instead of the original parking space information, otherwise the parking space information is continuously executed according to the existing parking space information. Stop the parking until the parking is completed.
- step A40 The automatic parking step is continued, and it is detected whether the automatic parking step is finished, or the process returns to step A30.
- step A30 the vehicle speed and the user command are also detected. When the vehicle speed is zero and the user issues an automatic parking instruction, step A30 is performed, otherwise the waiting for detecting step is performed.
- the waiting for detection step includes the following substeps:
- the effective parking range refers to whether the parking space is still in the identification range of the first parking space identification method or the second parking space recognition method.
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
一种停车位识别方法，包括如下步骤：获取车身姿态及位置；然后同时执行超声波车位识别算法以及AVM车位识别算法。利用超声波车位识别算法找到停车位后，在泊车进行时持续对停车位进行识别，若期间通过AVM车位识别算法识别到停车位信息，则将该停车位信息发送给自动泊车***以代替原来的停车位信息。
Description
本发明涉及自动泊车领域，特别涉及一种停车位识别方法及泊车方法及车载导航。
停车位识别是自动泊车***重要的感知单元，目前带有自动泊车功能的车辆，大部分采用纯超声波雷达车位识别或者全景车位识别的方案。
其中，纯超声波雷达车位识别方案，是通过长距雷达对车辆周边的障碍物信息进行检测，同时对检测到的数据进行相应的处理并输出停车位信息。该方案对车辆周边障碍物的依赖较大，车辆最终停放的效果以周边障碍物做参考，导致很多情况下使用自动泊车很难达到规范停车的要求。
全景车位识别方案，又称为AVM车位识别方案，是通过全景***对地面的车位线进行图像识别处理，识别可泊入的目标车位并输出相应的车位信息。对于有车位线的停车位，该方案可实现较为精准的停车效果，但是对于没有车位线或车位线比较特殊的停车位时，该方案将无法进行停车位识别。
发明内容
本发明为了解决上述技术问题，提供一种停车位识别方法及泊车方法。
一种停车位识别方法，包括如下步骤：
S10、采用航迹推算算法对车辆当前位置以及车身姿态进行定位；
S20、执行超声波车位识别算法对停车位进行识别以及绘制，获得第一停车位信息；
S30、将所述第一停车位信息发送给自动泊车***，同时执行AVM车位识别算法，在泊车进行时持续对车位进行识别，若期间识别到停车位信息，则将该停车位信息作为第二停车位信息，并将所述第二停车位信息发送给自动泊车***以代替第一停车位信息。
进一步的，所述步骤20包括如下子步骤：
S21、沿汽车行驶方向采集第一次搜索到障碍物的坐标点A1；
S22、继续采集第一次丢失障碍物的坐标点B1，以及采集第二次搜索到障碍物的坐标点C1；
S23、判断坐标点B1以及坐标点C1之间的距离是否符合停车标准，若是则执行第一停车位信息生成步骤，生成第一停车位信息，否则返回步骤S21。
进一步的，所述第一停车位信息包括如下子步骤：
S231、录入所获得的坐标点A1、坐标点B1、坐标点C1以及车辆当前坐标；
S232、将坐标点B1以及坐标点C1的坐标转换成相对车辆当前位置的坐标点B2以及坐标点C2；
S233、根据转换后的坐标构建出第一停车位信息。
进一步的，执行所述步骤S20时还同时执行入选步骤：
执行AVM车位识别算法，若识别到停车位信息，则将该停车位信息作为第二停车位信息，并将所述第二停车位信息发送给自动泊车***；忽略第一停车位信息。
进一步的，所述第一停车位信息和或第二停车位信息包括当前车辆位置坐标以及停车位边缘坐标。
另外，本发明还提供一种泊车方法，包括如下步骤：
A10、同时运行第一停车位识别算法和第二停车位识别算法，通过第一停车位识别算法获取低优先级的第一停车位信息，通过第二停车位算法获取高优先级的第二停车位信息；
A20、将所能获取到的优先级的较高的停车位信息发送给自动泊车***；
A30、持续执行第一停车位识别算法和第二停车位识别算法，并执行自动泊车步骤，期间当获取到更高优先级的停车位信息时，将更高优先级的停车位信息发送给自动泊车***用于自动泊车；
A40、检测自动泊车步骤是否结束，否者返回步骤A30。
进一步的，所述第一停车位识别算法为超声波停车位识别算法，所述第二停车位算法为AVM车位识别算法。
进一步的，所述步骤A30之前，还对车速以及用户指令进行检测，当车辆速度为零，且用户发出允许自动泊车指令时，执行步骤A30，否则执行等待检测步骤。
进一步的，所述等待检测步骤包括如下子步骤：
B10、利用航迹推算算法对车辆当前位置以及车身姿态进行持续定位；
B20、检测当前车辆位置是否超出有效停车范围，若否则返回步骤B10，若是则提示用户并返回步骤A10。
进一步的，当所述步骤A10无法搜索到有效车位信息时，返回车位搜索识别的提示。
本发明的停车位识别方法及泊车方法所起到的有益效果包括：
1、通过两种停车位识别算法的融合运用后，有效解决超声波车位识别的准确性问题，提高自动泊车的停车规范性，同时弥补AVM对于没有车位线或车位线比较特殊的停车位无法识别的缺陷。
2、由于兼顾了车位识别精度以及识别范围，因此大大提高了本发明自动泊车方法的 覆盖场景，也提高了自动泊车的有效性。
图1为本发明实施例1的停车位识别方法原理图。
图2为本发明实施例1的超声波车位识别算法原理示意图。
图3为本发明实施例2的停车位识别方法原理图。
图4为本发明实施例3的自动泊车方法原理图。
下面结合附图对本发明的较佳实施例进行详细阐述，以使本发明的优点和特征更易被本领域技术人员理解，从而对本发明的保护范围作出更为清楚的界定。
实施例1：
一种停车位识别方法，融合了AVM车位识别以及超声波雷达车位识别的两种方法对停车位进行更加有效而且精准的识别，同时解决了现有技术汇总超声波雷达车位识别的不精确，以及AVM车位识别范围小和识别局限的问题。具体如图1所示，包括如下步骤：
S10、采用航迹推算算法对车辆当前位置以及车身姿态进行定位，本实施例的航迹算法具体利用轮速传感器返回的轮速脉冲信息对车轮位移进行测量，根据采样周期内车辆位移的变化，计算车辆方位角增量和位置增量，得到车辆相对初始状态的位姿，并且经过持续运算，这可以获得车辆的航迹。
S20、执行超声波车位识别算法对停车位进行识别以及绘制，获得第一停车位信息。具体的如图2所示，其包括如下子步骤：
S21、启动算法后，汽车向前行驶，此时沿汽车行驶方向对车辆周围进行障碍物检车，通常情况下，可以通过用户选择操作来确定重点监测方向，如选择停车位在汽车左侧还是右侧等，确定后开始采集并记录第一次搜索到障碍物的坐标点A1；坐标的形式可以是二维坐标，也可以是角座标，通常情况下，采用的是二维坐标。
S22、在采集到坐标点A1后，算法继续采集障碍物坐标点，当连续监测到障碍物，并历尽关于坐标点A1的障碍物后，超声波雷达将失去该障碍物的检测，在失去该障碍物检测的那一刻所采集到的坐标点，称之为第一次丢失障碍物的坐标点B1。进而，继续对障碍物进行检测，当第二次搜索到障碍物时，记录该点的坐标，设置为坐标点C1。结合附图2，从整体角度看，坐标点B1和坐标点C1之间的位置即是可能为停车位。
S23、为了进一步判断停车位是否符合要求，及其宽度是否足够用于停车，可以判断坐标点B1以及坐标点C1之间的距离是否大于车宽的一定比例，若是则执行第一停车位信 息生成步骤，生成第一停车位信息，否则返回步骤S21。
其中，生成该第一停车位信息包括如下子步骤：
S231、录入所获得的坐标点A1、坐标点B1、坐标点C1以及车辆当前坐标。
S232、通过坐标的换算，将车辆当前坐标设置为原点，并且通过坐标差将坐标点B1以及坐标点C1的坐标转换成相对车辆当前位置的坐标点B2以及坐标点C2。优选的，可以将汽车后轴的中心点作为计算的原点，以保证泊车时更加准确。
S233、根据转换后的坐标构建出第一停车位信息，该第一停车位信息包括当前车辆位置坐标以及停车位边缘坐标，可以将停车位的形状进行构建绘画。
S30、在获得了第一停车位信息后，将第一停车位信息发送给自动泊车***，使自动泊车***可以进行自动泊车步骤，与此同时，当自动泊车进行时，控制执行AVM车位识别算法，持续对车位进行识别。为了提高识别的成功率，可以将识别范围设定在坐标点B1以及坐标点C1的范围附近，从而使AVM识别更加有目的性。
此期间若通过AVM车位识别算法识别到停车位信息，则将该停车位信息作为优先度较高的第二停车位信息。当第一停车位信息与第二停车位信息同时都被检测到时，将自动把第二停车位信息发送给自动泊车***，以代替第一停车位信息，自动泊车***采用较高优先级的第二停车位信息作为停车依据。
其中，与第一停车位信息的格式相同，第二停车位信息同样包括当前车辆位置坐标以及停车位边缘坐标。
实施例2：
作为实施例1的优化，本实施例与实施例1的区别在于：如图3所示，本实施例为了提高自动泊车的效率以及相应速度，本实施例在执行步骤S20时还同时，还会执行AVM车位识别算法。并且同时检测两种停车位信息获取情况。
若在AVM车位识别算法识别到停车位信息，则无论超声波车位识别算法是否识别到停车位信息，都将直接把AVM车位识别算法识别到停车位信息作为第二停车位信息，并将第二停车位信息发送给自动泊车***，忽略第一停车位信息。可以理解的，可以直接放弃执行步骤S30。
为了保证车位识别率的情况下，也可以持续执行超声波车位识别算法，从而持续获取第一停车位信息。
实施例2：
本实施例在实施例1与实施例2的基础上，还提供一种泊车方法，包括如下步骤：
A10、同时运行第一停车位识别算法和第二停车位识别算法，通过第一停车位识别算法获取低优先级的第一停车位信息，通过第二停车位算法获取高优先级的第二停车位信息。其中第一停车位识别算法可以使识别范围较广车位识别方法，用于初步确定车位，而第二停车位识别算法则可以与之配合地使用识别精度较高的方法。由于识别范围较广的停车位识别算法会先识别到停车位，进而提高第二停车位识别算法的识别速度。
在本实施例中，第一停车位识别算法可以但不仅限于为超声波停车位识别算法，第二停车位算法可以但仅限于为AVM车位识别算法。为了方便举例，后续将采用上述两种算法对本实施例所公开的技术方案进行阐述，如图4所示。
A20、为了提高自动泊车的精度，因此需要将所能获取到的优先级的较高的停车位信息发送给自动泊车***。即当同时获取到第一停车位信息和第二停车位信息的时候，将对优先级较高的第二停车位信息发送到自动泊车***，是***以第二停车位信息作为位置判断依据。而当无法识别到第二停车位信息是才使用第一停车位信息。
可以理解的，采用两种识别算法均无法搜索到有效车位信息时，将返回车位搜索识别的提示，同时可以根据用户的操作选择是否返回重新搜索车位。
A30、确定接受到停车位信息后，将根据所接收到的停车位信息进行自动泊车步骤，与此同时，仍然持续执行第一停车位识别算法和第二停车位识别算法，期间当获取到比当前采用的停车位信息更高优先级的停车位信息时，将更高优先级的停车位信息发送给自动泊车***，代替原有停车位信息，否则根据现有的停车位信息继续执行停车步骤，直到完成停车。
A40、持续进行自动泊车步骤，并检测自动泊车步骤是否结束，否者返回步骤A30。
另外，为了保证自动泊车步骤能够安全的介入，***需要接收到两个确定信息，其一为执行自动泊车步骤前，车辆处于停车状态，其二是获得了用户的允许。因此，步骤A30之前，还对车速以及用户指令进行检测，当车辆速度为零，且用户发出允许自动泊车指令时，执行步骤A30，否则执行等待检测步骤。
该等待检测步骤包括如下子步骤：
B10、利用航迹推算算法对车辆当前位置以及车身姿态进行持续定位。
B20、由于在用户选择自动泊车之前，车辆仍然会继续移动，因此需要持续检测当前车辆位置是否超出有效停车范围，若否则返回步骤B10，若是则提示用户并返回步骤A10。其中，有效停车范围指的是停车位是否仍处于第一停车位识别方法或第二停车位识别方法的识别范围。
上面结合附图对本发明的实施方式作了详细说明，但是本发明并不限于上述实施方 式，在本领域普通技术人员所具备的知识范围内，还可以在不脱离本发明宗旨的前提下作出各种变化。
Claims (11)
- 一种停车位识别方法，其特征在于：包括如下步骤：S10、对车辆当前位置以及车身姿态进行定位；S20、执行超声波车位识别算法对停车位进行识别以及绘制，获得第一停车位信息；S30、将所述第一停车位信息发送给自动泊车***，同时执行AVM车位识别算法，在泊车进行时持续对车位进行识别，若期间识别到停车位信息，则将该停车位信息作为第二停车位信息，并将所述第二停车位信息发送给自动泊车***以代替第一停车位信息。
- 根据权利要求1所述的停车位识别方法，其特征在于，所述步骤20包括如下子步骤：S21、沿汽车行驶方向采集第一次搜索到障碍物的坐标点A1；S22、继续采集第一次丢失障碍物的坐标点B1，以及采集第二次搜索到障碍物的坐标点C1；S23、判断坐标点B1以及坐标点C1之间的距离是否符合停车标准，若是则执行第一停车位信息生成步骤，生成第一停车位信息，否则返回步骤S21。
- 根据权利要求2所述的停车位识别方法，其特征在于，所述第一停车位信息包括如下子步骤：S231、录入所获得的坐标点A1、坐标点B1、坐标点C1以及车辆当前坐标；S232、将坐标点B1以及坐标点C1的坐标转换成相对车辆当前位置的坐标点B2以及坐标点C2；S233、根据转换后的坐标构建出第一停车位信息。
- 根据权利要求1所述的停车位识别方法及泊车方法，其特征在于，执行所述步骤S20时还同时执行入选步骤：执行AVM车位识别算法，若识别到停车位信息，则将该停车位信息作为第二停车位信息，并将所述第二停车位信息发送给自动泊车***；忽略第一停车位信息。
- 根据权利要求1所述的停车位识别方法，其特征在于，所述第一停车位信息和/或第二停车位信息包括当前车辆位置坐标以及停车位边缘坐标。
- 根据权利要求1所述的停车位识别方法，其特征在于，所述步骤S10中，采用航迹推算算法车辆当前位置以及车身姿态进行定位。
- 一种泊车方法，其特征在于，包括如下步骤：A10、同时运行第一停车位识别算法和第二停车位识别算法，通过第一停车位识别算法获取低优先级的第一停车位信息，通过第二停车位算法获取高优先级的第二停车位信息；A20、将所能获取到的优先级的较高的停车位信息发送给自动泊车***；A30、持续执行第一停车位识别算法和第二停车位识别算法，并执行自动泊车步骤，期间当获取到更高优先级的停车位信息时，将更高优先级的停车位信息发送给自动泊车***用于自动泊车；A40、检测自动泊车步骤是否结束，否者返回步骤A30。
- 根据权利要求7所述的泊车方法，其特征在于，所述第一停车位识别算法为超声波停车位识别算法，所述第二停车位算法为AVM车位识别算法。
- 根据权利要求7所述的泊车方法，其特征在于，所述步骤A30之前，还对车速以及用户指令进行检测，当车辆速度为零，且用户发出允许自动泊车指令时，执行步骤A30，否则执行等待检测步骤。
- 根据权利要求9所述的泊车方法，其特征在于，所述等待检测步骤包括如下子步骤：B10、利用航迹推算算法对车辆当前位置以及车身姿态进行持续定位；B20、检测当前车辆位置是否超出有效停车范围，若否则返回步骤B10，若是则提示用户并返回步骤A10。
- 根据权利要求7所述的泊车方法，其特征在于，当所述步骤A10无法搜索到有效车位信息时，返回车位搜索识别的提示。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810405667.8A CN108806308A (zh) | 2018-04-29 | 2018-04-29 | 一种停车位识别方法及泊车方法 |
| CN201810405667.8 | 2018-04-29 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019210599A1 true WO2019210599A1 (zh) | 2019-11-07 |
Family
ID=64093553
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/099191 WO2019210599A1 (zh) | 2018-04-29 | 2018-08-07 | 一种停车位识别方法及泊车方法 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108806308A (zh) |
| WO (1) | WO2019210599A1 (zh) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109800658B (zh) * | 2018-12-26 | 2023-05-26 | 中汽研（天津）汽车工程研究院有限公司 | 基于神经网络的泊车位类型在线识别与定位***及方法 |
| CN109703554B (zh) * | 2019-02-27 | 2020-11-24 | 湖北亿咖通科技有限公司 | 泊车车位确认方法及装置 |
| JP7190393B2 (ja) * | 2019-05-13 | 2022-12-15 | 本田技研工業株式会社 | 車両制御装置、車両管理装置、車両制御方法、およびプログラム |
| CN110444044B (zh) * | 2019-08-27 | 2022-07-12 | 纵目科技（上海）股份有限公司 | 基于超声波传感器的车辆位姿检测***、终端和存储介质 |
| CN111619553A (zh) * | 2020-05-26 | 2020-09-04 | 三一专用汽车有限责任公司 | 工程车辆和工程车辆的泊车控制方法 |
| CN112124303A (zh) * | 2020-09-27 | 2020-12-25 | 广州汽车集团股份有限公司 | 一种车位融合方法及*** |
| CN113482423A (zh) * | 2021-06-01 | 2021-10-08 | 上海追势科技有限公司 | 一种机械立体库位宽度推送方法 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103241239A (zh) * | 2013-04-27 | 2013-08-14 | 重庆邮电大学 | 一种自动泊车***车位识别方法 |
| CN103754219A (zh) * | 2014-02-21 | 2014-04-30 | 重庆邮电大学 | 一种多传感器信息融合的自动泊车*** |
| CN106427994A (zh) * | 2015-08-12 | 2017-02-22 | 现代自动车株式会社 | 停车辅助装置和方法 |
| KR20170055334A (ko) * | 2015-11-11 | 2017-05-19 | 엘지전자 주식회사 | 주차 난이도를 기초로 운전자를 지원하는 장치 및 방법 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101438941B1 (ko) * | 2012-12-11 | 2014-11-03 | 현대자동차주식회사 | 주차 지원 제어 장치 및 방법 |
| CN203318408U (zh) * | 2013-05-09 | 2013-12-04 | 北京工业大学 | 自动泊车*** |
| CN104354656B (zh) * | 2014-10-20 | 2016-08-24 | 同济大学 | 智能泊车***的障碍物检测与库位判别方法及实现*** |
| CN106671974B (zh) * | 2015-11-10 | 2019-09-20 | 新乡航空工业(集团)有限公司 | 一种用于智能泊车***的车位检测方法 |
| JP6724425B2 (ja) * | 2016-03-04 | 2020-07-15 | アイシン精機株式会社 | 駐車支援装置 |
| CN107521493A (zh) * | 2017-08-28 | 2017-12-29 | 苏州优达斯汽车科技有限公司 | 一种自动泊车*** |
-
2018
- 2018-04-29 CN CN201810405667.8A patent/CN108806308A/zh active Pending
- 2018-08-07 WO PCT/CN2018/099191 patent/WO2019210599A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103241239A (zh) * | 2013-04-27 | 2013-08-14 | 重庆邮电大学 | 一种自动泊车***车位识别方法 |
| CN103754219A (zh) * | 2014-02-21 | 2014-04-30 | 重庆邮电大学 | 一种多传感器信息融合的自动泊车*** |
| CN106427994A (zh) * | 2015-08-12 | 2017-02-22 | 现代自动车株式会社 | 停车辅助装置和方法 |
| KR20170055334A (ko) * | 2015-11-11 | 2017-05-19 | 엘지전자 주식회사 | 주차 난이도를 기초로 운전자를 지원하는 장치 및 방법 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108806308A (zh) | 2018-11-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019210599A1 (zh) | 一种停车位识别方法及泊车方法 | |
| CN109649384B (zh) | 一种泊车辅助方法 | |
| EP3929046B1 (en) | Vertical parking method | |
| CN109409202B (zh) | 基于动态感兴趣区域的鲁棒性车道线检测方法 | |
| CN109031323B (zh) | 基于超声波测距雷达的车位检测方法 | |
| US10606257B2 (en) | Automatic parking system and automatic parking method | |
| JP5782708B2 (ja) | 走行支援装置 | |
| CN105551302B (zh) | 辅助停车过程的方法以及停车辅助装置 | |
| CN108928343A (zh) | 一种全景融合自动泊车***及方法 | |
| CN108944915A (zh) | 一种自动泊车方法、***及计算机可读存储介质 | |
| CN111942372A (zh) | 一种自动泊车方法及*** | |
| US20180178780A1 (en) | Automatic parking method and system of vehicle | |
| CN111746504A (zh) | 记忆泊车方法、***、装置及计算机可读存储介质 | |
| JP2007030700A (ja) | 駐車支援装置 | |
| CN108149990A (zh) | 自动泊车车位探测方法和探测***以及车辆 | |
| CN112771591B (zh) | 用于评价运输工具的环境中的对象对运输工具的行驶机动动作的影响的方法 | |
| CN110803157A (zh) | 一种基于自动泊车的车位识别方法和*** | |
| CN103481820A (zh) | 一种带有安全监测的辅助泊车***及其方法 | |
| WO2019065970A1 (ja) | 車両用外界認識装置 | |
| CN112776797A (zh) | 泊出原始车位创建方法、***、车辆和存储介质 | |
| EP3342681A1 (en) | Automatic parking system and automatic parking method | |
| CN102436758B (zh) | 用于支持车辆停车过程的方法和设备 | |
| CN114495066A (zh) | 一种辅助倒车的方法 | |
| CN109559555B (zh) | 车位识别***及其识别方法 | |
| US10970870B2 (en) | Object detection apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18917565 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18917565 Country of ref document: EP Kind code of ref document: A1 |