A method for detecting and tracking the available parking status of storage locations in a set area

Abstract

The present invention discloses a method for detecting and tracking the parkable state of storage locations in a set area, including two scenarios: when the vehicle is in motion and when it stops at the position for monitoring the storage location state. In both scenarios, the monitoring and tracking of the storage location state in the target container area are carried out, reducing the limitations of the perception coverage range of the intelligent container truck and external occlusion, improving the real-time performance and accuracy of the parking space detection, ensuring that the vehicle can detect "parkable" storage locations or have more "parkable" storage locations to choose from under specific circumstances, and improving the operation efficiency and environmental compatibility of the intelligent container truck for parking.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous driving, and particularly relates to a method for detecting and tracking the parkable state of designated area berths. Background Art

[0002] In order to improve the operation efficiency, container ports are equipped with automated loading and unloading operations and dedicated parking berths. Currently, intelligent container trucks usually perform parking according to the target container area released by the cloud dispatching system. After the intelligent container truck arrives at the designated area and completes the berth search, it selects a parkable berth to complete the automatic parking action, and the gantry crane then performs the loading and unloading of containers according to the vehicle position.

[0003] However, in the current automatic berth search and parking scheme of intelligent container trucks, when all the berths of the intelligent container truck are occupied, the intelligent container truck cannot select a parkable berth. And in order not to affect the operation of other vehicles, the intelligent container truck cannot stay at the position of the container area entrance for a long time. It will terminate the automatic parking process after arriving at the designated position and wait for manual takeover. In addition, restricted by the perception range of the intelligent driving system and the occlusion of obstacles and vehicles on both sides of the berth, the intelligent container truck cannot directly obtain and update the real-time state of the berths in the container area.

[0004] The current berth state detection method cannot update the parkable state in real time: when the vehicle in the occupied berth that the intelligent container truck has passed by drives out, the actual state of the berth changes from "not parkable" to "parkable"; but in reality, restricted by the perception range of the intelligent driving system and the occlusion of obstacles and vehicles on both sides of the berth, the intelligent container truck cannot directly obtain and update the state of the berths in the container area, and the system still maintains the "not parkable" state judgment of the berth, resulting in a discrepancy between the available berths detected by the system and the actual situation.

[0005] Based on the above two scenarios, there are significant scenario limitations in the actual application of the current intelligent container truck's automatic parking function. Summary of the Invention

[0006] In view of the above problems, the main object of the present invention is to design a method for detecting and tracking the parkable state of designated area berths, so as to solve the problems that in the current automatic parking scheme of intelligent container trucks, due to the limitations of vehicle perception ability and system scheme, automatic parking cannot continue after all berths are occupied or the detected situation does not match the actual situation after the berth state changes.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A method for detecting and tracking the parkable status of storage locations in a set area, which is based on the intelligent driving system of a vehicle. The specific process of searching for the storage location includes the following steps:

[0009] S1: During the vehicle's travel, sense, detect, and track the status of the storage locations in the set area;

[0010] S2: The vehicle travels to the monitoring position of the storage location status in the set area;

[0011] S3: Monitor and track the parkable status of the storage locations in the set area;

[0012] S4: Update the information of the target parkable parking space;

[0013] S5: Update the starting position of vehicle parking;

[0014] S6: Plan the vehicle parking route and execute the parking.

[0015] As a further description of the present invention, before the above-mentioned S1, it also includes the limitation of the set area, specifically including:

[0016] S0: The vehicle obtains the target container area based on the scheduling system. The target container area is the set area, and the target container area includes several storage locations;

[0017] Based on the entrance side of the storage locations in the target container area, set the route for searching for the storage location. The route for searching for the storage location maintains a fixed distance from the edge on one side of the entrance of the storage location. The route for searching for the storage location includes the starting point position for searching for the storage location and the monitoring position of the storage location status.

[0018] As a further description of the present invention, the starting point position for searching for the storage location is set outside the outer edge of the starting storage location in the target container area, and it is the starting point of the entire process of the vehicle searching for the storage location;

[0019] The monitoring position of the storage location status is set outside the outer edge of the last storage location in the target container area, and it is the end point of the entire process of the vehicle searching for the storage location;

[0020] The vehicle's travel process is: drive along the route for searching for the storage location from the starting point position for searching for the storage location until the monitoring position of the storage location status.

[0021] As a further description of the present invention, during the travel process, the central axis of the vehicle is parallel to the route for searching for the storage location and maintains a fixed lateral deviation distance.

[0022] As a further description of the present invention, the status of the storage locations in the set area includes the parkable status and the non-parkable status. In S1, if a parkable status parking space is detected in the set area, perform the operations of S4 - S6; if no parkable status parking space is detected in the set area, perform the operations of S2 - S6.

[0023] As a further description of the present invention, when the set area includes multiple parkable state parking spaces, the intelligent driving system of the vehicle preferentially selects the parkable state parking space with a smaller parking space number as the target parking space for parking in.

[0024] As a further description of the present invention, the perception detection and tracking of the parking space state in S1 specifically include the following:

[0025] During the vehicle's traveling process, determine the position of each parking space relative to the vehicle itself according to the positioning information of the vehicle itself;

[0026] Obtain the movement direction and position information of the moving vehicles in the target box area;

[0027] Perform fusion tracking on the direction and position information of the moving vehicles;

[0028] Determine that the vehicle is a vehicle driving out of the parking space, and match the distance between the vehicle itself and the parking space position;

[0029] Determine that the parking space state changes from "not parkable" to "parkable".

[0030] As a further description of the present invention, the monitoring and tracking of the parkable state of the parking space in S3 specifically include the following:

[0031] When the vehicle is at the state monitoring position, determine the position of each parking space relative to the vehicle itself according to the positioning information of the vehicle itself;

[0032] Obtain the movement direction and position information of the moving vehicles in the target box area;

[0033] Perform fusion tracking on the direction and position information of the moving vehicles;

[0034] Determine that the vehicle is a vehicle driving out of the parking space, and match the distance between the vehicle itself and the parking space position;

[0035] Determine that the parking space state changes from "not parkable" to "parkable".

[0036] As a further description of the present invention, the method for determining that the vehicle is a vehicle driving out of the parking space is:

[0037] Under a continuous time duration, the movement direction of the vehicle is away from the parking space position, and there is a lateral position deviation between the side edge position of the vehicle and the side edge of the parking space, and the vehicle has driven out of the area affecting the vehicle's parking operation.

[0038] As a further description of the present invention, the area affecting the vehicle's parking operation includes a lateral area and a longitudinal area based on a single parking space;

[0039] The lateral area is the sum of half of the width of the parking space and the safety redundancy distance, and the longitudinal area is the sum of half of the width of the vehicle outside the route for searching for the parking space and the safety redundancy distance.

[0040] Compared with the prior art, the technical effect of the present invention is as follows:

[0041] The present invention provides a method for detecting and tracking the parkable state of storage locations in a set area, including two scenarios: when the vehicle is moving and when it stops at the monitoring position of the storage location state. The monitoring and tracking of the storage location state in the target container area in these two scenarios reduce the limitations of the intelligent container truck's perception coverage and external occlusion, improve the real-time performance and accuracy of the parking space detection, ensure that the vehicle can detect "parkable" storage locations or have more "parkable" storage locations to choose from under specific circumstances, and improve the operation efficiency and environmental compatibility of the intelligent container truck's parking. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the overall detection and tracking process of the present invention;

[0043] Figure 2 It is a schematic diagram of the overall scenario setting and requirements of the present invention;

[0044] Figure 3 It is a schematic diagram of the state change process of the storage location recognized by the intelligent driving system in the present invention;

[0045] Figure 4 It is a schematic diagram of the area affecting the vehicle's parking operation in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The present invention will be described in detail below with reference to the accompanying drawings:

[0047] A method for detecting and tracking the parkable state of storage locations in a set area, referring to Figure 1 as shown, this method is based on the intelligent driving system of the vehicle, and the specific process of finding the storage location includes the following steps:

[0048] S1: During the vehicle's movement, the perception detection and tracking of the storage location state in the set area;

[0049] S2: The vehicle moves to the monitoring position of the storage location state in the set area;

[0050] S3: The monitoring and tracking of the parkable state of the storage locations in the set area;

[0051] S4: Update the information of the target parking space that can be parked;

[0052] S5: Update the starting position of the vehicle's parking;

[0053] S6: Plan the vehicle's parking route and execute the parking.

[0054] It should be noted that in this embodiment, the storage location status in the set area includes a parkable status and a non-parkable status. In S1, if a parkable parking space is detected in the set area, the operations of S4 - S6 are executed; if no parkable parking space is detected in the set area, the operations of S2 - S6 are executed.

[0055] The monitoring and tracking process of the storage location status in the target container area is as follows:

[0056] Determine the position of each storage location relative to the vehicle based on the positioning information of the vehicle itself.

[0057] Obtain the movement direction and position information of the moving vehicles in the target container area.

[0058] Perform integrated tracking on the direction and position information of the moving vehicles.

[0059] Determine that the vehicle is a vehicle driving out of the storage location, and match the distance between the vehicle itself and the storage location.

[0060] Determine that the storage location status changes from "non-parkable" to "parkable".

[0061] It should also be noted that specifically, before S1 in this embodiment, it also includes the limitation of the set area, and the limitation content includes:

[0062] S0: The vehicle, based on the scheduling system and the wireless network, in cooperation with the vehicle's high-precision positioning system and combined with the high-precision map, obtains the target container area, which is the set area, and the target container area includes several storage locations;

[0063] Set a route for searching for storage locations on the entrance side of the storage locations in the target container area. The route for searching for storage locations maintains a fixed distance x from the edge on one side of the entrance of the storage location (i.e., in front of the target container area);

[0064] The route for searching for storage locations includes a starting point position for searching for storage locations and a monitoring position for the storage location status; the starting point position for searching for storage locations is set outside the outer edge of the starting storage location in the target container area (at a distance α from the outer edge of the storage location), and it is the starting point of the entire storage location search process for the vehicle; the monitoring position for the storage location status is set outside the outer edge of the last storage location in the target container area (at a distance y from the outer edge of the storage location), and it is the end point of the entire storage location search process for the vehicle; the vehicle's traveling process is: drive along the route for searching for storage locations from the starting point position for searching for storage locations until the monitoring position for the storage location status. During the traveling process, the vehicle is in a straightened state, and the vehicle's speed for searching for storage locations is controlled at v1 (±v 允许误差 ) km / h (calibrated value). In addition, the central axis of the intelligent container truck is parallel to the route for searching for storage locations and maintains a fixed lateral deviation distance β.

[0065] That is, during the entire parking process, the vehicle will detect and track whether the target bay position is parkable at a certain lateral distance of x meters from the target bay position, along the fixed direction of the route for searching the bay. According to the results of the bay detection and tracking, subsequent parking route planning and execution operations will be carried out. The specific scenario schematic diagram is as shown in Figure 2 shown.

[0066] Specifically, under normal circumstances, x is 20 - 50m, α is 20 - 300m, y is 13 - 15m, v1 is <20kph, v 允许误差 is <5kph, β is 0 - 0.5m, which are specifically determined according to the actual situation.

[0067] It should be noted that in this embodiment, the parkable states of the positions in the target bay include parkable and non - parkable:

[0068] Parkable: There are no obstacles between the position and the route for the intelligent straddle carrier to search for the bay.

[0069] Non - parkable: There are obstacles between the position and the route for the intelligent straddle carrier to search for the bay.

[0070] Specifically, in this embodiment, each operation step during the vehicle's bay - searching process is analyzed in detail, and the analysis content is disclosed as follows:

[0071] S1. During the vehicle's travel, the perception detection and tracking of the status of the positions in the set area;

[0072] Before the vehicle reaches the position for monitoring the bay status or the starting point for parking and entering the bay bound to the selected position (the distance between this point and the longitudinal center line of the target bay is set as z), based on the vehicle's own positioning information and combined with the information of the four corner points of the positions in the target bay, the position of each bay relative to the vehicle is determined. Combining the detection and fusion information of the lidar, camera, and millimeter - wave radar equipped on the vehicle itself, the direction and position information of the moving vehicle in the target bay are judged. When it is confirmed that the vehicle's moving direction is away from the bay position for a continuous time of t, and the lateral position deviation between the detected vehicle edge position and the side edge of a certain bay in the map (this side edge is the side edge in the driving direction of the intelligent driving vehicle) is within △d, and the vehicle has driven out of the area affecting the parking operation of the intelligent straddle carrier, it is determined that the vehicle has driven out of this bay, and the intelligent driving system of the vehicle will determine that the status of this bay changes from "non - parkable" to "parkable" state. The overall module processing flow is referred to Figure 3 shown.

[0073] It should be noted that under normal circumstances, z is 25 - 40m, t is 2 - 5s, △d is - 1.5 - 1.5m, which are specifically determined according to the actual situation.

[0074] Among them, as shown in Figure 4As shown in the figure, the areas affecting the parking operation of intelligent container trucks include the lateral area and the longitudinal area based on a single storage location, specifically:

[0075] Lateral area: On both sides of the longitudinal center line of the storage location, each with a width of L 泊车横向影响距离 = L 半个库位宽度 + L 横向安全冗余距离 (L 安全冗余距离 is a calibrated value). Usually, L 泊车横向影响距离 is 1.8 - 2.0 m, L 半个库位宽度 is 1.5 m, and L 横向安全冗余距离 is 0.3 - 0.5 m, which is determined according to the actual situation.

[0076] Longitudinal area: From the bottom of the storage location to L outside the route for the intelligent container truck to search for the storage location 泊车纵向影响距离 = L 半车宽 + L 纵向安全冗余距离 (L 安全冗余距离 is a calibrated value). Usually, L 泊车纵向影响距离 is 2.0 - 2.5 m, L 半车宽 is 1.5 m, and L 纵向安全冗余距离 is 0.5 - 1.0 m, which is determined according to the actual situation.

[0077] S2. The vehicle travels to the position for monitoring the status of the storage location in the set area;

[0078] During the process of the vehicle traveling along the route for searching for the storage location, if no "parkable" storage location is detected, the vehicle will travel to the position for monitoring the status of the storage location as the end point.

[0079] S3. Monitoring and tracking the parkable status of the storage locations in the set area (the vehicle travels to the position for monitoring the status of the storage location in the set area);

[0080] After reaching the position for monitoring the status of the storage location, based on the vehicle's own positioning position and combined with the information of the four corner points of the storage locations in the target container yard, determine the position of each storage location relative to the vehicle, and combine the detection fusion information of the lidar, camera, and millimeter-wave radar equipped on the vehicle to judge the direction and position information of the moving vehicle in the storage locations of the target container yard. When it is confirmed for a continuous time of t that the moving direction of the vehicle is away from the storage location, and the lateral position deviation between the detected vehicle edge position and the side edge of a certain storage location in the map (this side edge is the side edge in the driving direction of the intelligent driving vehicle) is within △d, and the vehicle has driven out of the area affecting the parking operation of the intelligent container truck, it is determined that the vehicle has driven out of this storage location, and the intelligent driving system of the vehicle will determine that the status of this storage location has changed from "not parkable" to "parkable".

[0081] It should be noted that t is 2 - 5 s, and △d is -1.5 - 1.5 m, which is determined according to the actual situation. In addition, the above areas affecting the parking operation of the intelligent container truck are for referenceFigure 3 as shown, and the detailed description in S1.

[0082] S4. Update the information of the target parking space that can be parked.

[0083] After detecting that there is a parking space available in the warehouse, the intelligent driving system of the vehicle preferentially selects the available parking space with a smaller warehouse space number as the target parking space for parking.

[0084] S5. Update the starting position of vehicle parking.

[0085] The intelligent driving system of the vehicle obtains the position information of the starting point for parking into the warehouse bound to the warehouse space from the map according to the target warehouse space number for parking. The vehicle will travel to this point and stop. The distance between this point and the longitudinal center line of the target warehouse space is set as z (calibrated value), and the central axis of the intelligent container truck is parallel to the route for searching the warehouse and maintains a fixed lateral deviation distance β, and the vehicle is in a straightened state.

[0086] Generally, z is 25 - 40m and β is 0 - 0.5m, which is specifically determined according to the actual situation.

[0087] S6: Plan the vehicle parking route and execute parking.

[0088] After the vehicle reaches the specified starting position for parking into the warehouse, it can start to plan the parking route and perform subsequent execution operations.

[0089] When the present invention cannot directly obtain the warehouse space status, a method for judging the parkable status of the warehouse space is carried out through the detection and fusion of the vehicle motion status in the target container area warehouse space; when all the warehouse spaces in the set target container area are occupied, the warehouse space status monitoring and tracking is carried out.

[0090] It should be noted that in this embodiment, an intelligent container truck is used for illustration, but it is also applicable to other vehicles in similar scenarios. In addition, this embodiment is also applicable to highly automated parking functions such as AVP with fixed warehouse spaces.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Other modifications or equivalent replacements made by those of ordinary skill in the art to the technical solutions of the present invention shall be covered within the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for detecting and tracking the parkable state of storage locations in a set area, characterized in that: This method is based on the intelligent driving system of the vehicle. The specific process of searching for a storage location includes the following steps: S1: During the vehicle's travel, sense, detect, and track the status of storage locations in the set area; S2: The vehicle travels to the monitoring position of the storage location status in the set area; S3: Monitor and track the parkable status of storage locations in the set area; S4: Update the information of the target parking space that can be parked; S5: Update the starting position of vehicle parking; S6: Plan the vehicle parking route and execute parking; Before S1, it also includes the limitation of the set area, specifically including: S0: The vehicle obtains the target container area based on the dispatching system. The target container area is the set area, and several storage locations are included in the target container area; Set the storage location search route based on the entrance side of the storage locations in the target container area. The storage location search route maintains a fixed distance from the edge on one side of the storage location entrance. The storage location search route includes the starting point position for storage location search and the monitoring position of the storage location status; The starting point position for storage location search is set outside the outer edge of the starting storage location in the target container area and is the starting point of the entire storage location search process for the vehicle; The monitoring position of the storage location status is set outside the outer edge of the last storage location in the target container area and is the end point of the entire storage location search process for the vehicle; The vehicle travel process is: drive along the storage location search route from the starting point position for storage location search until the monitoring position of the storage location status; During the travel process, the central axis of the vehicle is parallel to the storage location search route and maintains a fixed lateral deviation distance.

2. A method for detecting and tracking the parkable state of storage locations in a set area according to claim 1, characterized in that: The status of the storage locations in the set area includes parkable status and non-parkable status. In S1, if a parkable status parking space is detected in the set area, perform the operations of S4 - S6; if no parkable status parking space is detected in the set area, perform the operations of S2 - S6.

3. A method for detecting and tracking the parkable state of storage locations in a set area according to claim 2, characterized in that: When there are multiple parkable status parking spaces in the set area, the intelligent driving system of the vehicle preferentially selects the parkable status parking space with a smaller storage location number as the target parking space for parking.

4. A method for detecting and tracking the parkable state of storage locations in a set area according to claim 3, characterized in that: The sensing, detection, and tracking of the storage location status in S1 specifically include the following: During the vehicle's travel, determine the position of each storage location relative to the vehicle based on the vehicle's positioning information; Obtain the movement direction and position information of the moving vehicles in the target container area; Fusion track the direction and position information of the moving vehicles; Determine that the vehicle is a vehicle driving out of the storage location, and match the distance between the vehicle itself and the storage location position; Determine that the storage location status changes from "non-parkable" to "parkable".

5. A method for detecting and tracking the parkable state of storage locations in a set area according to claim 3, characterized in that: The monitoring and tracking of the parkable status of storage locations in S3 specifically include the following: When the vehicle is at the status monitoring position, determine the position of each storage location relative to the vehicle based on the vehicle's positioning information; Obtain the movement direction and position information of the moving vehicles in the target container area; Fusion track the direction and position information of the moving vehicles; Determine that the vehicle is a vehicle driving out of the storage location, and match the distance between the vehicle itself and the storage location position; Determine that the storage location status changes from "non-parkable" to "parkable".

6. A method for detecting and tracking the parkable state of storage locations in a set area according to claim 4 or 5, characterized in that: The method for determining that the vehicle is a vehicle driving out of the storage location is: Under a continuous time period, the movement direction of the vehicle is away from the storage location position, and there is a lateral position deviation between the side edge position of the vehicle and the side edge of the storage location, and the vehicle has driven out of the area affecting the vehicle parking operation.

7. A method for detecting and tracking the parkable state of storage locations in a set area according to claim 6, characterized in that: The area affecting the vehicle parking operation includes a lateral area and a longitudinal area based on a single storage location; The horizontal area is the sum of half of the width of the storage location and the safety redundancy distance, and the vertical area is the sum of half of the vehicle width beyond the route of searching for storage locations and the safety redundancy distance.

Images