TWI684962B - Parking spot detection system and method thereof - Google Patents

Parking spot detection system and method thereof Download PDF

Info

Publication number
TWI684962B
TWI684962B TW107147399A TW107147399A TWI684962B TW I684962 B TWI684962 B TW I684962B TW 107147399 A TW107147399 A TW 107147399A TW 107147399 A TW107147399 A TW 107147399A TW I684962 B TWI684962 B TW I684962B
Authority
TW
Taiwan
Prior art keywords
parking space
parking
geometric shape
data points
information
Prior art date
Application number
TW107147399A
Other languages
Chinese (zh)
Other versions
TW202025105A (en
Inventor
郝昱翔
盧玟翰
胡家睿
李則霖
廖育萱
Original Assignee
財團法人工業技術研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 財團法人工業技術研究院 filed Critical 財團法人工業技術研究院
Priority to TW107147399A priority Critical patent/TWI684962B/en
Priority to JP2019124962A priority patent/JP2020107308A/en
Application granted granted Critical
Publication of TWI684962B publication Critical patent/TWI684962B/en
Publication of TW202025105A publication Critical patent/TW202025105A/en

Links

Images

Landscapes

  • Traffic Control Systems (AREA)

Abstract

A parking spot detection system having a detecting unit, a map unit and a processing unit. The detecting unit obtains a plurality of location information and a plurality of strength information of a plurality of data points in an environment space. The map unit provides map information. The processing unit, connected to the detecting unit and the map unit, determines that whether a geometric shape formed by the adjacent data points belongs to a parking space according to the plurality of strength information and the plurality of location information of the plurality of data points. If the geometric shape is belonged to the parking space, the processing unit further incorporates the geometric shape and location belonged to the parking space into the map information according to the plurality of location information of the plurality of data points.

Description

停車位偵測系統及方法Parking space detection system and method

本發明是有關於一種偵測技術，且特別是有關於一種停車位偵測系統及其方法。The invention relates to a detection technology, and in particular to a parking space detection system and method.

在車輛進入一場域，並準備停車時，車輛的駕駛無法立即地知道場域是否有適合車輛的停車位。駕駛通常需在場域中繞行才能找到適合的車位。基此，如何能夠快速地提供駕駛合適停車位的資訊，是本領域人員所致力的課題。When the vehicle enters a field and is ready to stop, the driver of the vehicle cannot immediately know whether there is a parking space suitable for the vehicle in the field. Drivers usually need to go around the field to find a suitable parking space. Based on this, how to quickly provide information on driving suitable parking spaces is a subject dedicated to those skilled in the art.

本揭露提供一種停車位偵測系統及方法，將車位資訊整合於地圖當中，以利自動地找到合適的車位。The present disclosure provides a parking space detection system and method, which integrates parking space information into a map to automatically find a suitable parking space.

本揭露提供一種停車位偵測系統，具有偵測單元、圖資單元以及處理單元。偵測單元獲取在環境空間中多個資料點的位置資訊及強度資訊。圖資單元提供地圖資訊。處理單元，連接偵測單元及圖資單元，依據資料點的強度資訊辨識相鄰資料點所形成的幾何形狀是否屬於停車空間。若判斷幾何形狀屬於停車空間，處理單元還依據資料點的位置資訊將屬於停車空間的幾何形狀與位置整合於地圖資訊中。The present disclosure provides a parking space detection system, which has a detection unit, a picture information unit and a processing unit. The detection unit obtains position information and intensity information of multiple data points in the environmental space. The map information unit provides map information. The processing unit, the connection detection unit and the image data unit identify whether the geometric shape formed by adjacent data points belongs to the parking space according to the intensity information of the data points. If it is determined that the geometric shape belongs to the parking space, the processing unit also integrates the geometric shape and position belonging to the parking space into the map information according to the location information of the data point.

本揭露提供一種停車位偵測方法，具有下列步驟。獲取在環境空間中多個資料點的位置資訊及強度資訊；依據資料點的強度資訊辨識相鄰資料點所形成的幾何形狀是否屬於停車空間；以及若判斷幾何形狀屬於停車空間，依據資料點的位置資訊將屬於停車空間的幾何形狀與位置整合於地圖資訊中。The present disclosure provides a parking space detection method, which has the following steps. Obtain the location information and intensity information of multiple data points in the environmental space; identify whether the geometric shape formed by adjacent data points belongs to the parking space based on the intensity information of the data points; and if the geometric shape belongs to the parking space, determine The location information integrates the geometry and location of the parking space into the map information.

基於上述，本揭露的停車位偵測系統及方法能夠偵測停車空間，並將偵測到的停車空間整合於地圖中。Based on the above, the disclosed parking space detection system and method can detect parking spaces and integrate the detected parking spaces into a map.

為讓本發明的上述特徵和優點能更明顯易懂，下文特舉實施例，並配合所附圖式作詳細說明如下。In order to make the above-mentioned features and advantages of the present invention more obvious and understandable, the embodiments are specifically described below in conjunction with the accompanying drawings for detailed description as follows.

圖1繪示本揭露一實施例停車位偵測系統的系統示意圖。請參照圖1，停車位偵測系統100具有偵測單元110、圖資單元120以及處理單元130。FIG. 1 illustrates a system schematic diagram of a parking space detection system according to an embodiment of the disclosure. Please refer to FIG. 1. The parking space detection system 100 has a detection unit 110, a graphics unit 120 and a processing unit 130.

偵測單元110用以感測環境空間的各物件表面的相關資訊。特別是，偵測單元110會感測環境空間的位置資訊及強度資訊，並且以資料點的形式來表示對應各物件的位置資訊及強度資訊。舉例來說，在此實施例中，偵測單元110例如為光達感測器（Lidar Sensor）。光達感測器會對周遭發出多個光訊號（例如，雷射光），每一個光訊號會延伸至環境空間中的遠處與各物件的表面。藉此，光達感測器會基於每一個主動或被動激發的光訊號在空間中回傳而取得的訊號的強度，獲取相應於空間中每一個資料點的位置資訊及強度資訊。環境空間中的三維立體空間能夠基於相應每一個位置資訊的強度資訊而被建立。The detection unit 110 is used to sense the relevant information on the surface of each object in the environmental space. In particular, the detection unit 110 senses the position information and intensity information of the environment space, and represents the position information and intensity information corresponding to each object in the form of data points. For example, in this embodiment, the detection unit 110 is, for example, a Lidar sensor. The light sensor will emit multiple optical signals (for example, laser light) to the surroundings, and each optical signal will extend to the distance in the environmental space and the surface of each object. In this way, the light sensor will obtain the position information and intensity information corresponding to each data point in the space based on the intensity of the signal obtained by each active or passively excited light signal being returned in the space. The three-dimensional space in the environmental space can be established based on the intensity information corresponding to each position information.

圖資單元120提供地圖資訊，舉例來說，圖資單元120例如可以由各類型的導航裝置、車載裝置所實現，本揭露不限於此。在本揭露的實施例中，地圖資訊為高精度地圖（HD Map，High Definition Map）。高精度地圖能夠提供車輛當前的經緯度、高度及航向角等資訊，以提供更精細的行車與道路資訊。高精度地圖為本領域具有通常知識者所能夠理解的，於此不進行詳述。The map information unit 120 provides map information. For example, the map information unit 120 may be implemented by various types of navigation devices and in-vehicle devices, for example, and the disclosure is not limited thereto. In the embodiment of the present disclosure, the map information is a high-definition map (HD Map, High Definition Map). High-precision maps can provide information such as the vehicle's current latitude, longitude, altitude, and heading angle to provide more detailed driving and road information. High-precision maps can be understood by those with ordinary knowledge in the field, and will not be described in detail here.

處理單元130連接於偵測單元110以及圖資單元120，並用以執行停車位偵測系統100中的各類運算。處理單元130例如為，中央處理單元(Central Processing Unit，CPU)，或是其他可程式化之一般用途或特殊用途的微處理器(Microprocessor)、數位信號處理器(Digital Signal Processor，DSP)、可程式化控制器、特殊應用積體電路(Application Specific Integrated Circuit，ASIC)或其他類似元件或上述元件的組合，本揭露不限於此。The processing unit 130 is connected to the detection unit 110 and the graphics unit 120, and is used to perform various operations in the parking space detection system 100. The processing unit 130 is, for example, a central processing unit (Central Processing Unit, CPU), or other programmable general-purpose or special-purpose microprocessor (Microprocessor), digital signal processor (Digital Signal Processor, DSP), The programmable controller, the application specific integrated circuit (Application Specific Integrated Circuit, ASIC) or other similar components or the combination of the above components are not limited to this disclosure.

圖2繪示本揭露一實施例停車位偵測系統的情境示意圖。請參照圖2，圖2所繪示的情境相應於圖1的停車位偵測系統。在此實施例中，偵測單元110設置在車輛10的上方，以對環境空間進行偵測，然本揭露並不限制偵測單元110安設於車輛10上的位置。除此之外，在運行停車位偵測系統100時，車輛10可以處於移動狀態。也就是說，停車位偵測系統100會在移動時持續運行，並且重覆地偵測停車位20。FIG. 2 is a schematic diagram illustrating a situation of a parking space detection system according to an embodiment of the disclosure. Please refer to FIG. 2, the situation depicted in FIG. 2 corresponds to the parking space detection system of FIG. 1. In this embodiment, the detection unit 110 is disposed above the vehicle 10 to detect the environmental space. However, the disclosure does not limit the position where the detection unit 110 is installed on the vehicle 10. In addition, when the parking space detection system 100 is operated, the vehicle 10 may be in a moving state. In other words, the parking space detection system 100 will continue to operate while moving and repeatedly detect the parking space 20.

值得一提的是，在本揭露的一個應用情境當中，偵測單元110不僅會判斷周遭的環境是否有能夠容納車輛10的平面位置，更進一步地，偵測單元110還會在找到能夠容納車輛10的平面位置空間後，進一步判斷平面位置上的高度是否能夠容納車輛10。舉例來說，在圖2路邊停車的情境中，倘若樹的枝葉相當茂密，低過車輛10的高度，將使得車輛10停入停車位20時撞到枝葉。換句話說，樹的枝葉會對停車位20產生高度限制。此時，偵測單元110將會偵測出停車位20上方具有障礙物，進而排除停車位20。又或者是，在本揭露的其他情境中，車輛10駛入地下室，此時，偵測單元110會偵測容納車輛10的平面位置空間上方是否具有障礙物，例如，是否有因樓地板高、機械式車位等的高度限制造成車輛10無法停車的情形。藉此，以找到適應車輛10的停車位。詳細的過程請參照下方的說明。It is worth mentioning that in an application scenario disclosed in this disclosure, the detection unit 110 will not only determine whether the surrounding environment has a plane position capable of accommodating the vehicle 10, but furthermore, the detection unit 110 will also find that it can accommodate the vehicle After the plane position space of 10, it is further determined whether the height at the plane position can accommodate the vehicle 10. For example, in the scenario of roadside parking in FIG. 2, if the foliage of the tree is quite dense and lower than the height of the vehicle 10, the vehicle 10 will hit the foliage when it is parked in the parking space 20. In other words, the branches and leaves of the tree impose height restrictions on the parking space 20. At this time, the detection unit 110 will detect that there is an obstacle above the parking space 20, and then exclude the parking space 20. Or, in other scenarios of the present disclosure, the vehicle 10 is driving into the basement. At this time, the detection unit 110 will detect whether there is an obstacle above the space where the vehicle 10 is located in the plane, for example, whether there is a high floor, The height restrictions of mechanical parking spaces and the like cause the vehicle 10 to be unable to stop. In order to find a parking space suitable for the vehicle 10. Please refer to the description below for the detailed process.

圖3繪示本揭露一實施例停車位偵測方法的流程示意圖。請參照圖3，此停車位偵測方法至少適用於圖1及圖2實施例的停車位偵測系統100。以下將同時依據圖1至圖3說明本揭露停車位偵測系統100以及停車位偵測方法運行的過程。FIG. 3 is a schematic flowchart of a parking space detection method according to an embodiment of the disclosure. Please refer to FIG. 3, this parking space detection method is at least applicable to the parking space detection system 100 of the embodiments of FIGS. 1 and 2. Hereinafter, the operation process of the disclosed parking space detection system 100 and the parking space detection method will be described based on FIGS. 1 to 3 at the same time.

在步驟S310，由偵測單元110獲取在環境空間中多個資料點的位置資訊及強度資訊。In step S310, the detection unit 110 acquires position information and intensity information of multiple data points in the environmental space.

在步驟S320，由處理單元130依據資料點的強度資訊判斷相鄰資料點所形成的幾何形狀是否屬於停車空間。In step S320, the processing unit 130 determines whether the geometric shape formed by the adjacent data points belongs to the parking space according to the intensity information of the data points.

具體來說，偵測單元110會發送訊號至環境空間中，並接收在每一個位置資訊上所回傳的訊號。位置資訊為處理單元130依據環境空間所劃分的座標值，例如，處理單元130可以採用直角座標系記錄每一個資料點的位置資訊，然不以直角坐標系為限。任何能夠用來標示空間位置的方式皆可被應用於本揭露中。並且，由於回傳的訊號會因為接觸的障礙物的距離、材質而導致強度不一致。基此，處理單元130能夠進一步依據在每個位置資訊所回傳的強度資訊，進而建立相應環境空間的三維立體空間。舉例來說，在本揭露的一實施例中，位置資訊相應於柏油路以及位置資訊相應於路面標線時，兩者之間所回彈的訊號強度資訊差異約30%，然本揭露不限於此。Specifically, the detection unit 110 sends a signal to the environment space, and receives the signal returned on each position information. The position information is the coordinate value divided by the processing unit 130 according to the environmental space. For example, the processing unit 130 may use the rectangular coordinate system to record the position information of each data point, but not limited to the rectangular coordinate system. Any method that can be used to indicate the location of space can be used in this disclosure. In addition, due to the distance and material of the contacted obstacle, the strength of the returned signal will be inconsistent. Based on this, the processing unit 130 can further create a three-dimensional space corresponding to the environment space based on the intensity information returned at each position information. For example, in an embodiment of the present disclosure, when the position information corresponds to the asphalt road and the position information corresponds to the road marking, the difference in signal strength information rebounded between the two is about 30%, but the disclosure is not limited to this.

特別是，處理單元130會依據資料點的強度資訊而將強度資訊相似的相鄰資料點視為幾何形狀，此幾何形狀可能為三角形、四邊形或者任意不同形狀的圖形。然而，處理單元130僅在幾何形狀的形狀、大小符合停車空間的特徵與大小時，才會判斷幾何形狀是否屬於停車空間。具體的細節，將進一步於後方進行說明。In particular, the processing unit 130 considers adjacent data points with similar intensity information as geometric shapes according to the intensity information of the data points. The geometric shape may be a triangle, a quadrilateral, or a graph of any different shape. However, the processing unit 130 determines whether the geometric shape belongs to the parking space only when the shape and size of the geometric shape match the characteristics and size of the parking space. The specific details will be further described in the back.

在步驟S330，由處理單元130整合屬於停車空間的幾何形狀與位置資訊於地圖資訊中。In step S330, the processing unit 130 integrates the geometry and location information belonging to the parking space into the map information.

詳細來說，在本揭露的實施例中，地圖資訊為高精度地圖。並且，處理單元130依據位置資訊及強度資訊所建立的三維立體空間中，能夠明確的獲知在每個位置的障礙物。基此，處理單元130能夠整合屬於停車空間的幾何形狀，並在地圖資訊中標記此幾何形狀屬於停車空間。In detail, in the embodiment of the present disclosure, the map information is a high-precision map. In addition, in the three-dimensional space created by the processing unit 130 according to the position information and the intensity information, the obstacle at each position can be clearly learned. Based on this, the processing unit 130 can integrate the geometry belonging to the parking space, and mark this geometry in the map information as belonging to the parking space.

圖4繪示本揭露一實施例停車位偵測方法的流程示意圖。以下將通過圖4說明本揭露處理單元130依據資料點的強度資訊辨識相鄰資料點所形成的幾何形狀是否屬於停車空間的細節。FIG. 4 is a schematic flowchart of a parking space detection method according to an embodiment of the disclosure. The details of whether the geometric shape formed by the neighboring data points belong to the parking space according to the intensity information of the data points will be described by FIG. 4 in the following.

在步驟S410，由處理單元130依據地圖資訊辨識幾何形狀屬於移動物或非移動物，並濾除屬於移動物的幾何形狀。在本揭露的實施例中，地圖資訊是由圖資單元120所事先建立並提供的資訊。而處理單元130會依據接收到回傳訊號的位置資訊及強度資訊，進而建立三維立體空間。基此，通過比較地圖資訊以及三維立體空間，處理單元130會判斷同時存在地圖資訊以及三維立體空間中的幾何形狀是長期不動的，屬於非移動物。反之，處理單元130會將僅存在地圖資訊的幾何形狀，或者是僅存在三維立體空間中的幾何形狀，皆判斷為移動物。In step S410, the processing unit 130 recognizes that the geometric shape belongs to a moving object or a non-moving object according to the map information, and filters out the geometric shape belonging to the moving object. In the embodiment of the present disclosure, the map information is information created and provided by the map information unit 120 in advance. The processing unit 130 will create a three-dimensional space based on the position information and intensity information of the received return signal. Based on this, by comparing the map information and the three-dimensional space, the processing unit 130 will determine that the geometric shapes that exist in the map information and the three-dimensional space are immobile for a long time and belong to a non-moving object. On the contrary, the processing unit 130 determines that the geometric shape with only map information or the geometric shape with only three-dimensional space is a moving object.

除此之外，處理單元130也會整合一段時間中，在環境空間中重覆獲取的多個資料點的位置資訊及強度資訊，並且對整合後的位置資訊及強度資訊進行降階處理，以濾除屬於移動物的幾何形狀。具體來說，由於在運行停車位偵測系統100時，車輛10可以處於連續移動狀態。因此，在車輛10連續移動的過程中，偵測單元110會依據車輛10移動的方向持續的獲取資料點的位置資訊及強度資訊。此時，倘若處理單元130整合一段時間中，在環境空間中重覆獲取的多個資料點的位置資訊及強度資訊時，非移動物的資料點會連續重複的被獲取，而移動物的位置資訊及強度資訊僅會在一段時間中的一部分被獲取，或者是，會分布在環境空間中的大片面積中。因此，經過一段時間的整合，移動物的強度資訊會低於非移動物。倘若經過降階處理，處理單元130能夠濾除屬於移動物的幾何形狀。In addition, the processing unit 130 will also integrate the position information and intensity information of multiple data points repeatedly acquired in the environmental space for a period of time, and perform the order reduction processing on the integrated position information and intensity information to Filter out geometric shapes that belong to moving objects. Specifically, when the parking space detection system 100 is operated, the vehicle 10 may be in a continuous movement state. Therefore, during the continuous movement of the vehicle 10, the detection unit 110 continuously obtains the position information and strength information of the data point according to the direction in which the vehicle 10 moves. At this time, if the processing unit 130 integrates the position information and intensity information of multiple data points repeatedly acquired in the environmental space for a period of time, the data points of the non-moving objects will be continuously and repeatedly acquired, and the position of the moving objects Information and intensity information will only be obtained for a part of a period of time, or it will be distributed over a large area in the environmental space. Therefore, after a period of integration, the strength information of moving objects will be lower than that of non-moving objects. If the order reduction process is performed, the processing unit 130 can filter out the geometry belonging to the moving object.

由於停車空間的幾何形狀屬於非移動物，基此，倘若濾除屬於移動物的幾何形狀後，處理單元130能夠減少運算，並較精確地獲取停車空間。值得一提的是，在上述濾除屬於移動物的幾何形狀的方法中，處理單元130可以僅採用通過比較地圖資訊以及三維立體空間的方法，或者僅採用整合一段時間中，在環境空間中重覆獲取的多個資料點的位置資訊及強度資訊，並且對整合後的位置資訊及強度資訊進行降階處理的方法，處理單元130也可以同時採用前述兩種方法，本揭露不限於此。Since the geometry of the parking space belongs to non-moving objects, based on this, if the geometry belonging to the moving objects is filtered out, the processing unit 130 can reduce calculations and acquire the parking space more accurately. It is worth mentioning that, in the above method of filtering out geometric shapes belonging to moving objects, the processing unit 130 may only adopt a method of comparing map information and three-dimensional space, or only adopt integration for a period of time, and refocus in the environmental space. The method of overriding the acquired position information and intensity information of multiple data points, and performing a reduced-order processing on the integrated position information and intensity information. The processing unit 130 may also use the foregoing two methods at the same time, and the disclosure is not limited thereto.

在步驟S420，由處理單元130選取部分的資料點，並於被選取的資料點中，將在對應高度之一軸線上座標值最小者設為基準點，並移除在軸線上位於基準點一預設高度以上的其他資料點。從另一角度而言，由於軸線是對應於於高度，因此，被選取作為基準點的是位於地面上的資料點。在本揭露的實施例中，預設高度例如為，5公分、10公分、50公分、100公分等，本揭露並不限制預設高度的數值。倘若預設高度為5公分時，處理單元130會移除基準點上5公分以上的所有資料點，保留基準點往上5公分以內的所有資料點。藉此，處理單元130能夠盡量地僅保留地面的資料點，以減少處理單元130運算的負擔。並且，處理單元130也能較精確地獲取停車空間。須說明的是，處理單元130僅是基於在獲取幾何空間時，減少處理單元130運算造成的負擔與誤判，因而邏輯上地移除對預設高度以上的資料點，以簡化運算過程。然在處理單元130判斷是否存在幾何形狀之後，資料點仍然會被持續的被應用在判斷相應幾何形狀的高度或其他應用之中，本揭露不限於此。In step S420, the processing unit 130 selects a part of the data points, and among the selected data points, sets the smallest coordinate value on the axis of one of the corresponding heights as the reference point, and removes the pre-positioning on the reference point on the axis Set other data points above the height. From another perspective, since the axis corresponds to the height, the data points on the ground are selected as the reference points. In the embodiment of the present disclosure, the preset height is, for example, 5 cm, 10 cm, 50 cm, 100 cm, etc. The present disclosure does not limit the value of the preset height. If the preset height is 5 cm, the processing unit 130 will remove all data points above 5 cm on the reference point and keep all data points within 5 cm above the reference point. In this way, the processing unit 130 can keep only the data points on the ground as much as possible, so as to reduce the calculation load of the processing unit 130. In addition, the processing unit 130 can acquire the parking space more accurately. It should be noted that the processing unit 130 is only based on reducing the burden and misjudgment caused by the calculation of the processing unit 130 when acquiring the geometric space, so logically removing data points above a predetermined height to simplify the calculation process. However, after the processing unit 130 determines whether there is a geometric shape, the data points will still be continuously used to determine the height of the corresponding geometric shape or other applications, and the disclosure is not limited thereto.

值得一提的是，在本揭露不同的實施例中，步驟S410及步驟S420可以彼此獨立地被應用，或者是同時被應用於停車位偵測系統100及停車位偵測方法中。此外，在本揭露其他實施例中，亦可以選擇性地不採用步驟S410、S420，本揭露不限於此。It is worth mentioning that, in different embodiments of the present disclosure, step S410 and step S420 can be applied independently of each other, or both in the parking space detection system 100 and the parking space detection method. In addition, in other embodiments of the present disclosure, steps S410 and S420 may be optionally not used, and the present disclosure is not limited thereto.

在步驟S430，由處理單元130判斷幾何形狀是否符合停車標線特徵。在本揭露的實施例中，停車標線特徵例如為，平行四邊形的角、平行四邊形的邊、平行四邊形形狀的至少一個。舉例來說，倘若幾何形狀相同於平行四邊形的角，處理單元130會判斷幾何形狀符合停車標線特徵。倘若幾何形狀不符合任何標線特徵，則處理單元130會直接判斷此幾何形狀不屬於停車空間。In step S430, the processing unit 130 determines whether the geometric shape conforms to the parking marking feature. In the embodiment of the present disclosure, the parking marking feature is, for example, at least one of a corner of a parallelogram, a side of a parallelogram, and a shape of a parallelogram. For example, if the geometric shape is the same as the corner of the parallelogram, the processing unit 130 determines that the geometric shape conforms to the characteristics of the parking marking. If the geometric shape does not meet any of the marking characteristics, the processing unit 130 will directly determine that the geometric shape does not belong to the parking space.

在步驟S440，由處理單元130判斷幾何形狀所圍成的空間大小是否不小於預設空間大小。在本揭露的實施例中，預設空間大小例如為，寬2.5公尺，長6公尺。又或者是，預設空間大小也可以為寬4公尺，長12公尺等等。預設空間大小會依據法規或車輛10的大小等實務需求進行調整，本揭露不限於此。並且，處理單元130會判斷由幾何形狀所圍成的區域是否不小於預設空間大小。倘若幾何形狀所圍成的空間大小小於預設空間大小，處理單元130判斷幾何形狀不屬於停車空間。In step S440, the processing unit 130 determines whether the size of the space enclosed by the geometric shape is not less than the preset size. In the embodiment of the present disclosure, the preset space size is, for example, 2.5 meters wide and 6 meters long. Or, the preset space size can also be 4 meters wide, 12 meters long, and so on. The preset space size will be adjusted according to practical requirements such as regulations or the size of the vehicle 10, and the disclosure is not limited to this. Moreover, the processing unit 130 determines whether the area enclosed by the geometric shape is not smaller than the preset space size. If the space enclosed by the geometric shape is smaller than the preset space, the processing unit 130 determines that the geometric shape does not belong to the parking space.

在本揭露的其他實施例中，處理單元130會判斷符合停車標線特徵及預設空間大小的幾何形狀屬於停車空間。然而，在本實施例中，處理單元130會繼續執行下述步驟。In other embodiments of the present disclosure, the processing unit 130 determines that the geometric shape that conforms to the parking marking feature and the preset space size belongs to the parking space. However, in this embodiment, the processing unit 130 will continue to perform the following steps.

在步驟S450，由處理單元130依據資料點的位置資訊與強度資訊獲取相應幾何形狀上方的高度資訊，並判斷高度資訊是否不小於車輛高度。具體來說，在幾何形狀確立屬於停車空間後，處理單元130會由平面往上確認在相應幾何空間位置上的其他資料點的強度資訊，並依據強度資訊進而獲取在幾何形狀上方是否存在障礙物。倘若存在障礙物，則相應此幾何形狀的高度資訊即為平面到障礙物之間的高度（此亦即為此幾何形狀的有效高度）。車輛高度則為車輛10停車所需的高度，例如，1.5公尺。倘若處理單元130判斷高度資訊小於車輛高度，處理單元130判斷幾何形狀不屬於停車空間。In step S450, the processing unit 130 obtains height information above the corresponding geometric shape according to the position information and intensity information of the data point, and determines whether the height information is not less than the vehicle height. Specifically, after the geometric shape belongs to the parking space, the processing unit 130 will confirm the intensity information of other data points at the corresponding geometric space position from the plane upward, and then obtain whether there is an obstacle above the geometric shape according to the intensity information . If there is an obstacle, the height information of the corresponding geometric shape is the height from the plane to the obstacle (this is also the effective height of the geometric shape). The vehicle height is the height required for the vehicle 10 to stop, for example, 1.5 meters. If the processing unit 130 determines that the height information is less than the vehicle height, the processing unit 130 determines that the geometric shape does not belong to the parking space.

在步驟S460，由處理單元130判斷符合停車標線特徵、預設空間大小以及車輛高度的幾何形狀屬於停車空間。基此，處理單元130能夠進一步將屬於停車空間的幾何形狀整合並標記於地圖資訊中。In step S460, the processing unit 130 determines that the geometric shape that conforms to the parking mark feature, the preset space size, and the vehicle height belongs to the parking space. Based on this, the processing unit 130 can further integrate and mark the geometry belonging to the parking space in the map information.

圖5繪示本揭露再一實施例停車位偵測方法的流程示意圖。圖5的實施例能夠結合於圖1至圖4的停車位偵測系統100及停車位偵測方法。然而，為避免贅述，以下將採用圖1至圖3及圖5來說明此實施例停車位偵測方法運行的過程。FIG. 5 is a schematic flowchart of a parking space detection method according to another embodiment of the disclosure. The embodiment of FIG. 5 can be combined with the parking space detection system 100 and the parking space detection method of FIGS. 1 to 4. However, in order to avoid redundant description, the operation process of the parking space detection method of this embodiment will be described below using FIG. 1 to FIG. 3 and FIG. 5.

請參照圖5，在步驟S510，由偵測單元110獲取在環境空間中多個資料點的位置資訊及強度資訊。在步驟S520，由處理單元130依據資料點的強度資訊辨識相鄰資料點所形成的幾何形狀是否屬於停車空間。在步驟S530，由處理單元130整合屬於停車空間的幾何形狀於地圖資訊中。步驟S510至步驟S530相同於步驟S310至步驟S330，於此不再贅述。Referring to FIG. 5, in step S510, the detection unit 110 acquires position information and intensity information of multiple data points in the environmental space. In step S520, the processing unit 130 recognizes whether the geometric shape formed by the adjacent data points belongs to the parking space according to the intensity information of the data points. In step S530, the processing unit 130 integrates the geometry belonging to the parking space into the map information. Steps S510 to S530 are the same as steps S310 to S330, and will not be repeated here.

在步驟S540，由處理單元130獲取車輛10的當前位置，並依據當前位置以及整合後的地圖資訊規劃車輛10的停車路徑。也就是說，處理單元130可以進一步判斷車輛10所在的當前位置，並依此執行路徑規劃，以產生停車路徑。並且，處理單元130更可以進一步在整合後的地圖資訊中提供導航。In step S540, the processing unit 130 acquires the current position of the vehicle 10, and plans the parking path of the vehicle 10 according to the current position and the integrated map information. In other words, the processing unit 130 can further determine the current position of the vehicle 10 and execute path planning accordingly to generate a parking path. Moreover, the processing unit 130 can further provide navigation in the integrated map information.

不僅如此，在此實施例中，停車位偵測系統100更具有停車單元。停車單元例如為，由連接於處理單元130的控制器所控制的馬達、方向盤等機械所共同實作。停車單元為本領域技術人員所能輕易理解的，於此不再詳述。Not only that, in this embodiment, the parking space detection system 100 further has a parking unit. The parking unit is implemented, for example, by a machine such as a motor or a steering wheel controlled by a controller connected to the processing unit 130. The parking unit is easily understood by those skilled in the art and will not be described in detail here.

在步驟S550，由停車單元依據停車路徑執行自動或輔助停車。停車單元會依據停車路徑將車輛10移動至停車空間。並且，為了確保安全，停車單元會再次判斷停車空間中是否存在其他車輛或障礙物。倘若停車空間中不具有車輛或障礙物，即執行自動或輔助停車。In step S550, the parking unit performs automatic or assisted parking according to the parking path. The parking unit moves the vehicle 10 to the parking space according to the parking path. And, to ensure safety, the parking unit will again determine whether there are other vehicles or obstacles in the parking space. If there are no vehicles or obstacles in the parking space, automatic or assisted parking is performed.

基此，停車位偵測系統100及停車位偵測方法不僅能夠將停車空間整合於地圖資訊上。更進一步地，停車位偵測系統100及停車位偵測方法更能夠提供停車位導航，並進一步被應用於自動或輔助停車中。Based on this, the parking space detection system 100 and the parking space detection method can not only integrate the parking space on the map information. Furthermore, the parking space detection system 100 and the parking space detection method can further provide parking space navigation, and are further applied to automatic or auxiliary parking.

雖然本發明已以實施例揭露如上，然其並非用以限定本發明，任何所屬技術領域中具有通常知識者，在不脫離本發明的精神和範圍內，當可作些許的更動與潤飾，故本發明的保護範圍當視後附的申請專利範圍所界定者為準。Although the present invention has been disclosed as above with examples, it is not intended to limit the present invention. Any person with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be subject to the scope defined in the appended patent application.

10‧‧‧車輛 20‧‧‧停車位 100‧‧‧停車位偵測系統 110‧‧‧偵測單元 120‧‧‧圖資單元 130‧‧‧處理單元 S310~S330、S410~S460、S510~550‧‧‧步驟10‧‧‧Vehicle 20‧‧‧ parking 100‧‧‧Parking space detection system 110‧‧‧detection unit 120‧‧‧Purchase unit 130‧‧‧Processing unit S310~S330, S410~S460, S510~550

圖1繪示本揭露一實施例停車位偵測系統的系統示意圖。 圖2繪示本揭露一實施例停車位偵測系統的情境示意圖。 圖3繪示本揭露一實施例停車位偵測方法的流程示意圖。 圖4繪示本揭露一實施例停車位偵測方法的流程示意圖。 圖5繪示本揭露再一實施例停車位偵測方法的流程示意圖。FIG. 1 illustrates a system schematic diagram of a parking space detection system according to an embodiment of the disclosure. FIG. 2 is a schematic diagram illustrating a situation of a parking space detection system according to an embodiment of the disclosure. FIG. 3 is a schematic flowchart of a parking space detection method according to an embodiment of the disclosure. FIG. 4 is a schematic flowchart of a parking space detection method according to an embodiment of the disclosure. FIG. 5 is a schematic flowchart of a parking space detection method according to another embodiment of the disclosure.

S310~S330‧‧‧步驟 S310~S330‧‧‧Step

Claims (16)

一種停車位偵測系統，包括： 一偵測單元，獲取在環境空間中多個資料點的位置資訊及強度資訊； 一圖資單元，提供地圖資訊；以及 一處理單元，連接該偵測單元及該圖資單元，依據該些資料點的強度資訊判斷相鄰資料點所形成的幾何形狀是否屬於停車空間，其中， 若判斷該幾何形狀屬於該停車空間，該處理單元還依據該些資料點的位置資訊將屬於該停車空間的該幾何形狀與位置整合於該地圖資訊中。A parking space detection system includes: a detection unit to obtain position information and intensity information of multiple data points in an environmental space; a map information unit to provide map information; and a processing unit to connect the detection unit and The image data unit determines whether the geometric shape formed by the adjacent data points belongs to the parking space according to the intensity information of the data points, and if the geometric shape is judged to belong to the parking space, the processing unit also depends on the data points The location information integrates the geometry and location belonging to the parking space into the map information. 如申請專利範圍第1項所述的停車位偵測系統，其中， 該處理單元還判斷該幾何形狀是否符合停車標線特徵， 該處理單元還判斷該幾何形狀所圍成的空間大小是否不小於預設空間大小， 該處理單元還判斷符合該停車標線特徵及該預設空間大小的該幾何形狀屬於該停車空間。The parking space detection system as described in item 1 of the patent application scope, wherein the processing unit further determines whether the geometric shape conforms to the characteristics of the parking marking, and the processing unit further determines whether the size of the space enclosed by the geometric shape is not less than The preset space size, the processing unit also determines that the geometric shape that matches the parking marking feature and the preset space size belongs to the parking space. 如申請專利範圍第2項所述的停車位偵測系統，其中，該停車標線特徵為平行四邊形的角、平行四邊形的邊、平行四邊形形狀的至少一個。The parking space detection system as described in Item 2 of the patent application range, wherein the parking marking feature is at least one of a parallelogram corner, a parallelogram edge, and a parallelogram shape. 如申請專利範圍第2項所述的停車位偵測系統，其中，該處理單元還依據該些資料點的該強度資訊與該位置資訊獲取相應該幾何形狀上方的高度資訊，並判斷該高度資訊是否不小於一車輛高度， 該處理單元還判斷符合該停車標線特徵、該預設空間大小以及該車輛高度的該幾何形狀屬於該停車空間。The parking space detection system as described in item 2 of the patent application scope, wherein the processing unit further obtains height information corresponding to the geometric shape according to the intensity information and the position information of the data points, and determines the height information Whether the height is not less than a vehicle height, the processing unit also determines whether the geometric shape that conforms to the parking mark feature, the preset space size, and the vehicle height belongs to the parking space. 如申請專利範圍第1項所述的停車位偵測系統，其中，該處理單元還依據地圖資訊辨識該幾何形狀屬於移動物或非移動物，並濾除屬於該移動物的該幾何形狀。The parking space detection system according to item 1 of the patent application scope, wherein the processing unit further recognizes that the geometric shape belongs to a moving object or a non-moving object according to map information, and filters out the geometric shape belonging to the moving object. 如申請專利範圍第1項所述的停車位偵測系統，其中，該處理單元還選取部分的該些資料點，於該些被選取的資料點中，將在對應高度之一軸線上座標值最小者設為基準點，並移除在該軸線上位於該基準點一預設高度以上的其他資料點。The parking space detection system as described in item 1 of the patent application scope, wherein the processing unit further selects some of the data points, and among the selected data points, the coordinate value on one axis corresponding to the height is the smallest Set as a reference point, and remove other data points on the axis that are above a predetermined height of the reference point. 如申請專利範圍第1項所述的停車位偵測系統，其中，該處理單元還獲取一車輛的當前位置，並依據整合後的該地圖資訊與該車輛的該當前位置規劃該車輛的停車路徑。The parking space detection system as described in item 1 of the patent scope, wherein the processing unit also obtains the current position of a vehicle and plans the parking path of the vehicle based on the integrated map information and the current position of the vehicle . 如申請專利範圍第7項所述的停車位偵測系統，更包括： 停車單元，連接於該處理單元，依據該停車路徑執行停車。The parking space detection system as described in item 7 of the patent application scope further includes: a parking unit connected to the processing unit and performing parking according to the parking path. 一種停車位偵測方法，包括： 獲取在環境空間中多個資料點的位置資訊及強度資訊； 依據該些資料點的強度資訊判斷相鄰資料點所形成的幾何形狀是否屬於停車空間；以及 若判斷該幾何形狀屬於該停車空間，依據該些資料點的位置資訊將屬於該停車空間的該幾何形狀與位置整合於一地圖資訊中。A parking space detection method includes: acquiring position information and intensity information of multiple data points in an environmental space; judging whether the geometric shape formed by adjacent data points belongs to the parking space based on the intensity information of the data points; and if It is determined that the geometric shape belongs to the parking space, and the geometric shape and position belonging to the parking space are integrated into a map information according to the position information of the data points. 如申請專利範圍第9項所述的停車位偵測方法，其中，於依據該些資料點的該強度資訊辨識該些相鄰資料點所形成的該幾何形狀是否屬於停車空間的步驟中，還包括： 判斷該幾何形狀是否符合停車標線特徵； 判斷該幾何形狀所圍成的空間大小是否不小於預設空間大小；以及 判斷符合該停車標線特徵及該預設空間大小的該幾何形狀屬於該停車空間。The parking space detection method as described in item 9 of the patent application scope, wherein in the step of identifying whether the geometric shape formed by the adjacent data points belongs to the parking space based on the intensity information of the data points, Including: judging whether the geometric shape conforms to the characteristics of the parking marking; judging whether the size of the space enclosed by the geometric shape is not smaller than the preset space size; and judging that the geometric shape conforming to the characteristics of the parking marking and the predetermined space size belongs to The parking space. 如申請專利範圍第10項所述的停車位偵測方法，其中，該停車標線特徵為平行四邊形的角、平行四邊形的邊、平行四邊形形狀的至少一個。The parking space detection method as described in item 10 of the patent application range, wherein the parking marking feature is at least one of a corner of a parallelogram, a side of a parallelogram, and a shape of a parallelogram. 如申請專利範圍第10項所述的停車位偵測方法，更包括： 依據該些資料點的該強度資訊與該位置資訊獲取相應該幾何形狀上方的高度資訊，並判斷該高度資訊是否不小於一車輛高度；以及 判斷符合該停車標線特徵、該預設空間大小以及該車輛高度的該幾何形狀屬於該停車空間。The parking space detection method as described in item 10 of the patent application scope further includes: obtaining height information corresponding to the geometric shape according to the intensity information and the position information of the data points, and determining whether the height information is not less than A vehicle height; and the geometric shape determined to conform to the parking mark feature, the preset space size, and the vehicle height belongs to the parking space. 如申請專利範圍第9項所述的停車位偵測方法，更包括： 依據該地圖資訊辨識該幾何形狀屬於移動物或非移動物，並濾除屬於該移動物的該幾何形狀。The parking space detection method as described in item 9 of the patent application scope further includes: identifying that the geometric shape belongs to a moving object or a non-moving object based on the map information, and filtering out the geometric shape belonging to the moving object. 如申請專利範圍第9項所述的停車位偵測方法，更包括： 選取部分的該些資料點；以及 於該些被選取的資料點中，將在對應高度之一軸線上座標值最小者設為基準點，並移除在該軸線上位於該基準點一預設高度以上的其他資料點。The parking space detection method as described in item 9 of the patent application scope further includes: selecting some of the data points; and among the selected data points, the smallest coordinate value on one axis corresponding to the height is set It is a reference point, and other data points that are above a predetermined height above the reference point on the axis are removed. 如申請專利範圍第9項所述的停車位偵測方法，更包括： 獲取一車輛的當前位置，並依據整合後的該地圖資訊與該車輛的該當前位置規劃該車輛的停車路徑。The parking space detection method as described in item 9 of the patent application scope further includes: acquiring the current position of a vehicle, and planning the parking path of the vehicle based on the integrated map information and the current position of the vehicle. 如申請專利範圍第15項所述的停車位偵測方法，更包括： 依據該停車路徑執行停車。The parking space detection method as described in item 15 of the patent application scope further includes: Carrying out parking according to the parking path.
TW107147399A 2018-12-27 2018-12-27 Parking spot detection system and method thereof TWI684962B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
TW107147399A TWI684962B (en) 2018-12-27 2018-12-27 Parking spot detection system and method thereof
JP2019124962A JP2020107308A (en) 2018-12-27 2019-07-04 Parking spot detection system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW107147399A TWI684962B (en) 2018-12-27 2018-12-27 Parking spot detection system and method thereof

Publications (2)

Publication Number Publication Date
TWI684962B true TWI684962B (en) 2020-02-11
TW202025105A TW202025105A (en) 2020-07-01

Family

ID=70413309

Family Applications (1)

Application Number Title Priority Date Filing Date
TW107147399A TWI684962B (en) 2018-12-27 2018-12-27 Parking spot detection system and method thereof

Country Status (1)

Country Link
TW (1) TWI684962B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103473950A (en) * 2012-06-06 2013-12-25 刘鉵 Parking lot parking space monitoring method
US20150339924A1 (en) * 2014-05-21 2015-11-26 Douglas J. Cook Parking space occupancy
US20160221495A1 (en) * 2013-11-06 2016-08-04 Frazier Cunningham, III Parking signaling system
US20170043808A1 (en) * 2015-08-12 2017-02-16 Hyundai Motor Company Parking assist apparatus and method
TW201826231A (en) * 2017-01-04 2018-07-16 鴻海精密工業股份有限公司 System used in car park and car park using the system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103473950A (en) * 2012-06-06 2013-12-25 刘鉵 Parking lot parking space monitoring method
US20160221495A1 (en) * 2013-11-06 2016-08-04 Frazier Cunningham, III Parking signaling system
US20150339924A1 (en) * 2014-05-21 2015-11-26 Douglas J. Cook Parking space occupancy
US20170043808A1 (en) * 2015-08-12 2017-02-16 Hyundai Motor Company Parking assist apparatus and method
TW201826231A (en) * 2017-01-04 2018-07-16 鴻海精密工業股份有限公司 System used in car park and car park using the system

Also Published As

Publication number Publication date
TW202025105A (en) 2020-07-01

Similar Documents

Publication Publication Date Title
AU2020202527B2 (en) Using wheel orientation to determine future heading
EP2429877B1 (en) Camera system for use in vehicle parking
US10409288B2 (en) Systems and methods for projecting a location of a nearby object into a map according to a camera image
JP6190758B2 (en) Object recognition device and vehicle
CN108463690B (en) Evaluating U-turn feasibility
CN103874931B (en) For the method and apparatus of the position of the object in the environment for asking for vehicle
US8410919B2 (en) Driving support apparatus
JP4847051B2 (en) Vehicle surrounding monitoring method and system
JP6780611B2 (en) Autonomous driving device
US11691619B2 (en) Automatic parking system and automatic parking method
CN111028534B (en) Parking space detection method and device
TW201704067A (en) Collision avoidance method, computer program product for said collision avoidance method and collision avoidance system
JP6926976B2 (en) Parking assistance device and computer program
JP2024023319A (en) Detection of emergency vehicle
JP4869745B2 (en) Depression angle calculation device, depression angle calculation method, depression angle calculation program, and image processing apparatus
JP7398196B2 (en) Parking support device and parking support method
CN108680157B (en) Method, device and terminal for planning obstacle detection area
JP2019197356A (en) Automatic driving system
US20150046015A1 (en) Method and Device for Assisting in Returning a Vehicle After Leaving a Roadway
JP2009211624A (en) Driving support device, driving support method, and computer program
JP2015069289A (en) Lane recognition device
JP2007264717A (en) Lane deviation decision device, lane deviation prevention device and lane follow-up supporting device
US10769948B2 (en) Parking spot detection system and method thereof
JP6776842B2 (en) Parking support method and parking support device
US20220172490A1 (en) Image processing apparatus, vehicle control apparatus, method, and program