Vehicle control method and apparatus

Abstract

Disclosed in the embodiments of the present disclosure are a vehicle control method and apparatus. The method relates to the fields of autonomous driving technology and autonomous vehicles, and comprises: in response to receiving first spatial information of a first workstation, acquiring first sensing information of a first vehicle, wherein the first spatial information is generated on the basis of initial map boundary data, and both the first sensing information and the initial map boundary data are used for determining the boundary of a working area; adjusting the first spatial information on the basis of the first sensing information, so as to obtain second spatial information of the first workstation, wherein the degree of matching between the second spatial information and the boundary of the working area is greater than the degree of matching between the first spatial information and the boundary of the working area; and generating a first control instruction on the basis of the second spatial information, wherein the first control instruction is used for controlling the first vehicle to travel to the first workstation. The present disclosure solves the technical problem in the prior art of the control precision of an autonomous mining truck being reduced and thus affecting the working efficiency due to map updates being affected by network conditions.

Description

Vehicle control method and device

[0001] Cross-reference to related applications

[0002] The present disclosure claims priority to a Chinese patent application with the priority number 202410635258.2, the title of which is "Vehicle control method and device", which was filed on May 21, 2024 with the China Patent Office, the entire content of which is incorporated herein by reference. TECHNICAL FIELD

[0003] The present disclosure relates to the field of automatic driving technology and unmanned vehicles, in particular, to a vehicle control method and device. BACKGROUND

[0004] In a mine operation scene, in order to facilitate efficient cooperation between a shovel and an unmanned mine truck, a loading position for parking the unmanned mine truck needs to be generated before the unmanned mine truck parks. The current loading position generation method is based on the latest high-precision map. After the vehicle collects the map information and uploads it to the cloud server, the cloud server generates the latest high-precision map and issues it to all vehicle ends. Since the map update depends on the network of the operation area, if the network is poor, the high-precision map cannot be updated in time, resulting in poor precision of the generated loading position, and further reducing the control precision of the unmanned mine truck and affecting the operation efficiency.

[0005] At present, there is no effective solution to the above problems. SUMMARY

[0006] The embodiments of the present disclosure provide a vehicle control method and device to at least solve the technical problem that the map update is affected by the network in the related art, resulting in reduced control precision of the unmanned mine truck and affecting the operation efficiency.

[0007] According to an aspect of an embodiment of the present disclosure, a vehicle control method is provided, including: in response to receiving first space information of a first operation position, obtaining first perception information of a first vehicle, wherein the first space information is generated based on initial map boundary data, and the first perception information and the initial map boundary data are both used to determine a boundary of an operation area; adjusting the first space information based on the first perception information to obtain second space information of the first operation position, wherein a matching degree between the second space information and the boundary of the operation area is greater than a matching degree between the first space information and the boundary of the operation area; and generating a first control instruction based on the second space information, wherein the first control instruction is used to control the first vehicle to travel to the first operation position.

[0008] According to another aspect of the embodiments of the present disclosure, a vehicle control device is also provided, comprising: an obtaining module configured to obtain first perception information of a first vehicle in response to receiving first space information of a first work site, wherein the first space information is generated based on initial map boundary data, and the first perception information and the initial map boundary data are both used to determine a boundary of a work area; an adjusting module configured to adjust the first space information based on the first perception information to obtain second space information of the first work site, wherein a matching degree between the second space information and the boundary of the work area is greater than a matching degree between the first space information and the boundary of the work area; and a generating module configured to generate a first control instruction based on the second space information, wherein the first control instruction is used to control the first vehicle to travel to the first work site.

[0009] According to another aspect of the embodiments of the present disclosure, an automatic driving vehicle is also provided, comprising: a memory storing an executable program; and a processor configured to run the program, wherein the program performs the method in the various embodiments of the present disclosure when running.

[0010] According to another aspect of the embodiments of the present disclosure, a computer readable storage medium is also provided, comprising a stored executable program, wherein the computer readable storage medium controls a device where the computer readable storage medium is located to perform the method in the various embodiments of the present disclosure when the executable program runs.

[0011] According to another aspect of the embodiments of the present disclosure, a computer program product is also provided, comprising a computer program, which, when executed by a processor, implements the method in the various embodiments of the present disclosure.

[0012] According to another aspect of the embodiments of the present disclosure, a computer program product is also provided, comprising a non-volatile computer readable storage medium storing a computer program, which, when executed by a processor, implements the method in the various embodiments of the present disclosure.

[0013] According to another aspect of the embodiments of the present disclosure, a computer program is also provided, which, when executed by a processor, implements the method in the various embodiments of the present disclosure.

[0014] In the embodiments of the present disclosure, in response to receiving the first space information of the first work site, the first perception information of the first vehicle is acquired, the first space information is adjusted based on the first perception information to obtain second space information of the first work site, and the first control instruction is generated based on the second space information to control the first vehicle to travel to the first work site. It is easy to note that after the first space information of the first work site is generated based on the initial map boundary data, the first space information can be adjusted by real-time perception of the first vehicle to obtain second space information with higher precision, and the first vehicle is controlled to travel based on the second space information so that the first vehicle is parked in place, thereby realizing the technical effect of generating a high-precision loading site, improving the control precision of the unmanned mine truck, and further improving the operation efficiency, thereby solving the technical problem that in the related art, map updating is affected by the network, resulting in reduced control precision of the unmanned mine truck and affecting the operation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0015] The drawings described herein are used to provide further understanding of the present disclosure, and form a part of the present application. The illustrative embodiments of the present disclosure and their descriptions serve to explain the present disclosure, and do not constitute an improper limitation on the present disclosure. In the drawings:

[0016] FIG. 1 is a flowchart of a vehicle control method according to an embodiment of the present disclosure;

[0017] FIG. 2 is a schematic diagram of an actual boundary, a perception boundary and a map boundary of an optional work area according to an embodiment of the present disclosure;

[0018] FIG. 3 is a schematic diagram of adjusting first space information according to an embodiment of the present disclosure;

[0019] FIG. 4 is a schematic diagram of an optional first work site and a to-be-worked site according to an embodiment of the present disclosure;

[0020] FIG. 5 is a schematic diagram of a vehicle control device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0021] In order for those skilled in the art to better understand the present disclosure, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present disclosure.

[0022] It should be noted that the terms "first", "second", and the like in the description and claims of the present disclosure and the above-described accompanying drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0023] Embodiment 1

[0024] According to an embodiment of the present disclosure, a vehicle control method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0025] FIG. 1 is a flowchart of a vehicle control method according to an embodiment of the present disclosure, as shown in FIG. 1, the method comprises the following steps:

[0026] In step S102, in response to receiving first space information of a first work site, first perception information of a first vehicle is obtained, wherein the first space information is generated based on initial map boundary data, and the first perception information and the initial map boundary data are both used to determine the boundary of the work area.

[0027] The first vehicle described above can be an unmanned vehicle that needs to be parked at a work site generated based on a high-precision map. The specific type of the first vehicle can be determined according to the actual work scene, for example, in a mine work scene, the first vehicle can be an unmanned mine truck, an unmanned excavator, etc., and in other work scenes, the first vehicle can also be an unmanned tractor, an unmanned harvester, etc., but not limited thereto.

[0028] The first work site described above can be a position where the first vehicle needs to be parked to cooperate with other vehicles for work, in a mine work scene, the first work site can be a loading site where the mine truck is parked, but not limited thereto. The first space information can refer to the spatial information of the first work site in the entire work area, which usually includes the position information of the first work site in the work area, and optionally, in order to facilitate the first vehicle to park and stop, the first space information can also include the orientation information of the first work site relative to the current position of the first vehicle.

[0029] The initial map boundary data described above can be used to determine the data of the boundary of the work area, which can be composed of a series of position coordinates of points. By connecting the series of points, the boundary of the work area can be obtained. Considering that the first work site is usually located on the boundary of the work area, the spatial information of the first work site can be generated based on the initial map boundary data, thereby obtaining the first spatial information described above. The initial map boundary data here can be the map boundary data in the latest high-precision map after the cloud server updates the latest high-precision map; or the map boundary data provided by other vehicles after the work is completed. Optionally, the map boundary data can be data generated by other vehicles by sensing the boundary of the work area through sensing sensors, or can be map boundary data received by other vehicles from the high-precision map shared by the cloud server; or can be map boundary data in the historical high-precision map when the cloud server does not update the high-precision map.

[0030] It should be noted that since multiple vehicles will be parked at the first work site in turn, if the first spatial information is generated by the multiple vehicles themselves, the boundaries of the work area determined by the multiple vehicles will be inconsistent, which will lead to inconsistent work sites generated by the multiple vehicles, and the inconsistency in the transverse direction will cause the problem that the vehicles cannot enter or exit the first work site due to blockage. Therefore, in the present disclosure, the first spatial information of the first work site can be generated by the cloud server, other vehicles, or work equipment cooperating with the first vehicle.

[0031] In an optional embodiment, the first vehicle can communicate with the cloud server through a network, and the cloud server can generate the first spatial information based on the initial map boundary data and deliver it to the first vehicle. In another optional embodiment, the first vehicle can communicate with other vehicles (for example, the vehicle that has completed the work and driven away from the first work site) through a network, and the other vehicles can forward the first spatial information generated based on the initial map boundary data to the first vehicle. In yet another optional embodiment, the first vehicle can communicate with the work equipment (for example, in a mining scene, the first vehicle can be an unmanned mining truck, and the work equipment can be a excavator) that needs to cooperate with the work through a network, and the work equipment can generate the first spatial information based on the initial map boundary data sent by the cloud server and send it to the first vehicle. Optionally, in the present disclosure, the way of sending the first spatial information by the work equipment is preferred.

[0032] As shown in FIG. 2, the first spatial information can be generated by the excavator based on the map boundary data in the latest high-precision map. At this time, the map boundary determined based on the map boundary data is represented by a dashed line, and the actual boundary of the work area is represented by a solid line. As can be seen from the figure, the difference between the map boundary and the actual boundary is far, and the accuracy of the first spatial information determined based on the map boundary is low.

[0033] To overcome the above problems, in an optional embodiment, the work area, in particular the boundary of the work area, can be perceived by the perception sensor on the first vehicle to obtain the first perception information described above, so that the first vehicle can determine the perceived boundary of the work area by processing the first perception information. As shown in FIG. 2, the perceived boundary (indicated by a dotted line) is closer to the actual boundary relative to the map boundary, and therefore, the accuracy of determining the first work site based on the perceived boundary is higher than the accuracy of determining the first work site based on the map boundary.

[0034] In step S104, the first spatial information is adjusted based on the first perception information to obtain second spatial information of the first work site, wherein the matching degree between the second spatial information and the boundary of the work area is greater than the matching degree between the first spatial information and the boundary of the work area.

[0035] It should be noted that, in order to ensure that the first vehicle can be smoothly and safely parked on the first work site, the boundary of the work area needs to intrude into the interior of the first work site, and the position of the boundary of the work area in the interior of the first work site can be set according to the actual work situation.

[0036] On this basis, it can be known that the first work site and the work area do not completely coincide, and therefore, the matching degree of the first work site between the boundaries of the work area can be determined by the coincidence degree between the first work site and the work area, for example, the higher the coincidence degree, the lower the matching degree of the first work site between the boundaries of the work area, and the lower the coincidence degree, the higher the matching degree of the first work site between the boundaries of the work area.

[0037] In an optional embodiment, the perceived boundary of the work area can be determined based on the first perception information, and the overlap degree of the perceived area and the first work site can be determined based on the first spatial information and the perceived boundary, if the overlap degree is greater than a pre-set overlap threshold, it indicates that the gap between the first spatial information and the actual spatial information of the first work site is large, and the first spatial information needs to be adjusted, for example, the first spatial information can be adjusted towards the direction of the perceived boundary, until, based on the new spatial information and the perceived boundary, it is determined that the overlap degree of the perceived area and the first work site is less than or equal to the overlap threshold, at this time, the new spatial information is the second spatial information described above.

[0038] Optionally, since the boundary of the work area needs to intrude into the interior of the first work site, the matching degree of the first work site between the boundaries of the work area can be determined by the intrusion degree of the boundary of the work area into the first work site, for example, the higher the intrusion degree, the higher the matching degree of the first work site between the boundaries of the work area, and the lower the intrusion degree, the lower the matching degree of the first work site between the boundaries of the work area.

[0039] As shown in FIG. 2, since the second space information is obtained by adjusting the first space information based on the first perception information, that is, the second space information is determined based on the perception boundary of the first vehicle, and the first space information is determined based on the initial map boundary data, that is, the first space information is determined based on the map boundary, and the gap between the map boundary and the actual boundary is greater than the gap between the perception boundary and the actual boundary, therefore, the matching degree calculated based on the second space information is greater than the matching degree calculated based on the first space information.

[0040] In an optional embodiment, the perception boundary of the work area can be determined based on the first perception information, and the intrusion degree of the perception boundary intruding the first work site based on the first space information, if the intrusion degree is less than a pre-set intrusion threshold, or the perception boundary does not intrude the first work site, it indicates that the gap between the first space information and the actual space information of the first work site is large, and the first space information needs to be adjusted, for example, the first space information can be adjusted towards the direction of the perception boundary, until the intrusion degree of the perception boundary intruding the first work site based on the new space information is greater than or equal to the intrusion threshold, at this time, the new space information is the second space information.

[0041] As shown in FIG. 3, the map boundary determined based on the initial map boundary data is shown as a dashed line, the perception boundary determined based on the first perception information is shown as a dotted line, and the actual boundary of the work area is shown by a solid line. The first work site corresponding to the first space information is shown by a dashed box, and the intrusion degree of the perception boundary intruding the first work site shown by the solid line box is greater than or equal to a pre-set threshold. Therefore, it can be determined that the perception boundary does not intrude the first work site, and the first work site needs to be moved to the left until it coincides with the solid line box, at this time, it can be determined that the space information corresponding to the solid line box is the second space information.

[0042] In step S106, the first control instruction is generated based on the second space information, wherein the first control instruction is used to control the first vehicle to drive to the first work site.

[0043] The first control instruction described above can be an instruction for controlling the first vehicle to automatically drive, and can be generated by a controller of the first vehicle, or can be generated by a cloud server. In the present disclosure, the first control instruction can be generated by the controller of the first vehicle.

[0044] In an optional embodiment, the driving trajectory of the first vehicle can be generated based on the second space information and the current space information of the first vehicle, and the first control instruction can be generated based on the driving trajectory to control the first vehicle to drive to the first work site. The first vehicle can accurately stop at the first work site, and the orientation of the first vehicle is the same as the orientation information in the second space information.

[0045] By the above steps provided by the embodiments of the present disclosure, in response to receiving the first spatial information of the first work site, the first perception information of the first vehicle is acquired, the first spatial information is adjusted based on the first perception information, the second spatial information of the first work site is obtained, the first control instruction is generated based on the second spatial information to control the first vehicle to travel to the first work site. It is easy to note that after the first spatial information of the first work site is generated based on the initial map boundary data, the first spatial information can be adjusted by the real-time perception of the first vehicle to obtain second spatial information with higher precision, and the first vehicle is controlled to travel based on the second spatial information, so that the first vehicle is parked in place, thereby realizing the generation of high-precision loading sites, improving the control precision of the unmanned mine truck, and further improving the technical effect of the operation efficiency, thereby solving the technical problem that in the related art, the map update is affected by the network, resulting in a decrease in the control precision of the unmanned mine truck and affecting the operation efficiency.

[0046] In the above embodiments of the present disclosure, adjusting the first spatial information based on the first perception information to obtain the second spatial information of the first work site comprises: determining, based on the first spatial information and the first perception information, a first intrusion degree of a boundary of a work area corresponding to the first perception information with respect to the first work site corresponding to the first spatial information; in response to the first intrusion degree being less than a preset threshold, adjusting the first spatial information to obtain the second spatial information, so that a second intrusion degree of the boundary of the work area corresponding to the first perception information with respect to the first work site corresponding to the second spatial information is greater than or equal to the preset threshold.

[0047] The first intrusion degree and the second intrusion degree described above can refer to the depth of the boundary of the work area and the first work site, and the higher the intrusion degree, the higher the matching degree of the first work site between the boundaries of the work area, and the lower the intrusion degree, the lower the matching degree of the first work site between the boundaries of the work area.

[0048] The preset threshold described above can be a preset intrusion degree as a standard. If the intrusion degree determined based on a spatial information is greater than or equal to the threshold, it indicates that the spatial information is accurate and does not need to be adjusted; if the intrusion degree determined based on a spatial information is less than the threshold, it indicates that the spatial information is inaccurate and needs to be adjusted.

[0049] In an optional embodiment, the first spatial information can be adjusted in an iterative manner. Optionally, a first invasion degree of a boundary of the work area corresponding to the first perception information relative to the first work site corresponding to the first spatial information can be determined based on the first spatial information and the first perception information. If the first invasion degree is greater than or equal to a preset threshold, it indicates that the first spatial information does not need to be adjusted, and the iterative process can be ended, and the first spatial information can be directly taken as the second spatial information. If the first invasion degree is less than the preset threshold, it indicates that the first spatial information needs to be adjusted, and the first spatial information can be adjusted according to a preset adjustment step to obtain a new spatial information. Then, a new invasion degree of the boundary of the work area corresponding to the first perception information relative to the first work site corresponding to the new spatial information is further determined. If the new invasion degree is still less than the preset threshold, the first spatial information can be adjusted according to the preset adjustment step, and the iterative process can be continued, and the above operations can be repeated until the new invasion degree is greater than or equal to the preset threshold. In the case where the new invasion degree is greater than or equal to the preset threshold, the new spatial information corresponding to the new invasion degree can be determined as the second spatial information.

[0050] By setting the preset threshold, the adjustment process of the first spatial information can be quantized, and the first spatial information can be adjusted without distinction, which can waste computing resources and affect work efficiency.

[0051] In the above embodiments of the present disclosure, the first spatial information at least includes first position information and a first orientation of the first work site. The first spatial information is adjusted to obtain the second spatial information, including: adjusting the first position information according to the first orientation to obtain second position information; and obtaining the second spatial information based on the first orientation and the second position information.

[0052] The first position information described above can refer to the position coordinates of the first work site in a pre-constructed coordinate system. Considering that the shape of the first work site is usually rectangular, the first position information can include the position coordinates of the four vertices of the first work site. In order to reduce the amount of data, the content of the first position information can be simplified, for example, including the position coordinates of the vertices on the same diagonal, and for example, including the position coordinates of a vertex, as well as the length and width, but not limited thereto. In order to ensure that the first vehicle is parked on the first work site and can be directly matched with other vehicles for work, the first spatial information can further include the first orientation. As shown in FIG. 2, in a mine work scene, a shovel and an unmanned mine vehicle work together. The mine vehicle can be parked facing the shovel, so the first orientation is opposite to the orientation of the shovel, or the mine vehicle can be parked facing away from the shovel, so the first orientation is the same as the orientation of the shovel.

[0053] In an optional embodiment, as shown in FIG. 3, the first work position determined based on the map boundary and the first work position determined based on the perception boundary only change in the lateral direction, that is, only the position information changes, and the orientation is the same. Therefore, the first position information can be adjusted in the direction of the first orientation to obtain second position information, and the first orientation can be directly used as the orientation information in the second spatial information.

[0054] In the above-mentioned embodiments of the present disclosure, the method further comprises: in response to receiving third spatial information of a to-be-worked position, generating a second control instruction based on the third spatial information, wherein the to-be-worked position is used to park the first vehicle in the case that the first vehicle is parked on the first work position, and the second control instruction is used to control the first vehicle to travel to the to-be-worked position; and in response to the first vehicle traveling to the to-be-worked position, obtaining the first spatial information.

[0055] In actual work, the same work position will be used to park the autonomous vehicle in turn. In order to avoid waiting for too many vehicles and affecting the work efficiency, a to-be-worked position can be generated at the same time as the work position is generated, and all vehicles need to enter the to-be-worked position to wait before entering the work position.

[0056] The third spatial information mentioned above can also be generated based on the initial map boundary data, and the difference from the first spatial information is the position. As shown in FIG. 4, for the same work equipment, two loading positions (loading position 1 and loading position 2) can be generated, and two to-be-loaded positions (to-be-loaded position 1 and to-be-loaded position 2) can also be generated. The two to-be-loaded positions are farther away from the boundary of the work area, and the interval between the to-be-loaded position and the loading position can be adjusted according to the actual work efficiency requirement.

[0057] The second control instruction mentioned above can be an instruction for controlling the second vehicle to automatically travel, and can be generated by the controller of the first vehicle or by the cloud server. In the present disclosure, the second control instruction can be generated by the controller of the first vehicle.

[0058] In an optional embodiment, the farther the first vehicle is from the boundary of the work area, the lower the accuracy of the first perception information obtained. In order to avoid wasting computing resources, a travel trajectory of the first vehicle can be generated based on the third spatial information and the current spatial information of the first vehicle, and a second control instruction can be generated based on the travel trajectory to control the first vehicle to travel to the to-be-worked position. In the case that the first vehicle is accurately parked at the to-be-worked position, the first spatial information can be obtained, the first perception information can be collected through the perception sensor on the vehicle, and then the first spatial information can be adjusted in the above-mentioned manner, and the first vehicle can be controlled to be accurately parked at the first work position, and the orientation of the first vehicle is the same as the orientation information in the second spatial information.

[0059] In the above embodiments of the present disclosure, the first spatial information is spatial information generated based on a second orientation of the work equipment and third position information, the work equipment is used to work in cooperation with the vehicle at the first work site, and the third position information is determined based on initial map data sent by the cloud server, and the initial map data includes initial map boundary data.

[0060] The second orientation described above can be an orientation of the work equipment, and the third position information described above can be a specific position where the work equipment is parked outside the work area. Optionally, the second orientation and the third position information are determined based on operation instructions received by the work equipment, or determined based on third perception information of the work equipment, or determined based on work requirements of the work equipment. The work requirements here can be a work plane where the work equipment is located, work planning, terrain, work range, etc., but are not limited thereto.

[0061] The first spatial information can be generated by the work equipment and sent to the first vehicle. In an optional embodiment, the driver of the work equipment can first adjust the orientation of the work equipment (for example, place the excavator arm), and then mark the specific position and the current orientation of the work equipment on the initial map data by the driver, so that the work equipment can obtain the second orientation and the third position information. In another optional embodiment, the third perception information can be collected by the sensor on the work equipment, and the second orientation and the third position information can be determined in combination with the initial map data. In yet another optional embodiment, the second orientation and the third position information can be automatically determined based on the work requirements of the work equipment in combination with the initial map data, so as to realize automatic operation.

[0062] After the second orientation and the third position information are determined, the first spatial information of the first work site and the third spatial information of the to-be-worked site are generated based on the work site generation method in the related art. Optionally, the first spatial information of the first work site and the third spatial information of the to-be-worked site can be generated by a vector algorithm in combination with work parameters (such as arm length, rotation angle, etc.) of the work equipment and the degree of boundary intrusion of the work area to the work site, but are not limited thereto, and other ways can be used to realize, for example, a neural network model.

[0063] In the above embodiments of the present disclosure, the initial map data is obtained by updating historical map data of the work area based on perception map boundary data sent by the second vehicle, wherein the perception map boundary data is generated based on second perception information of the second vehicle after the second vehicle drives out of the second work site.

[0064] The second vehicle is also an unmanned vehicle, and the second vehicle and the first vehicle both need to cooperate with other vehicles for operation. The difference between the first vehicle and the second vehicle is that the first vehicle is about to enter the first operation position for operation, and the second vehicle is about to leave the second operation position after operation is completed.

[0065] In an optional embodiment, the initial map data can be synchronized to all vehicles, including the first vehicle, the second vehicle and the operation equipment, by the cloud server. When the second vehicle leaves the second operation position after operation is completed, the second vehicle can obtain second sensing information by sensing the boundary of the operation area through the sensors on the second vehicle, and then generate sensing map boundary data based on the second sensing information, and upload the sensing map boundary data to the cloud server by the second vehicle. After receiving the sensing map boundary data, the cloud server can update the historical map data based on the sensing map boundary data to obtain the initial map data, that is, update the historical map boundary data in the historical map data by using the sensing map boundary data to obtain the initial map boundary data in the initial map data. The specific updating process can be directly replacing the historical map boundary data with the sensing map boundary data, or comparing the historical map boundary data with the sensing map boundary data and adjusting the historical map boundary data according to the comparison result, but is not limited thereto.

[0066] In the above-mentioned embodiments of the present disclosure, the initial map data is obtained by replacing the set of boundary points between the historical endpoints in the historical map data based on the sensing map boundary data, and the historical endpoint is an endpoint searched in the search tree corresponding to the historical map data based on the sensing endpoint in the sensing map boundary data, wherein the distance between the historical endpoint and the sensing endpoint is less than the distance between other boundary points in the search tree and the sensing endpoint, and the other boundary points are used to represent the boundary points in the search tree other than the historical endpoint.

[0067] The above-mentioned search tree can refer to a data structure for storing boundary point position information and supporting efficient search, insertion and deletion operations. The search tree adopts a tree structure, and each node in the tree contains a value and a pointer to its child node, for example, it can be a binary search tree, but is not limited thereto. The above-mentioned sensing endpoint can be an endpoint representing the two ends of the boundary of the operation area, and the above-mentioned historical endpoint can be the nearest neighbor endpoint of the sensing endpoint in the search tree.

[0068] In an optional embodiment, the cloud server can construct a search tree for the historical map data, and then search for the nearest neighbor endpoints of the two endpoints of the sensing boundary in the search tree by using the two endpoints of the sensing boundary in the sensing map boundary data, and also search for the serial numbers, that is, obtain the above-mentioned historical endpoints. Then, all the point sets between the nearest neighbor points in the search tree are replaced by the point data in the sensing map boundary data to obtain the initial map data.

[0069] In the above-mentioned embodiments of the present disclosure, the perception map boundary data is data obtained by boundary extraction on a grid map corresponding to the work area based on target categories of different grids in the grid map, the target categories are determined based on target probabilities of the grids corresponding to multiple categories, and the target probabilities are probabilities obtained by updating historical probabilities of the grids using recognition probabilities of the grids. The recognition probabilities are probabilities of the grids corresponding to multiple categories obtained by recognizing the current grid map using a deep learning model. The current grid map is a result of grid processing on the second perception information. The grid map corresponding to the work area is a result of accumulation on multiple historical grid maps. The historical probabilities are cumulative results of probabilities of the grids corresponding to multiple categories obtained by recognizing the multiple historical grid maps using the deep learning model.

[0070] The above-mentioned grid map corresponding to the work area can be a local grid map of a fixed size maintained by the second vehicle. The fixed size can be set according to actual needs (for example, the size of the work area). For example, the fixed size can be 100*100 m or 1000*1000 m, but is not limited thereto.

[0071] In an optional embodiment, after obtaining the second perception information, the second vehicle can convert the second perception data into a fixed-size semantic grid to obtain a current grid map (i.e., a single-frame semantic grid dem). The fixed size can be set according to actual needs (for example, the size of the work area). For example, the fixed size can be 0.2*0.2 m or 0.5*0.5 m, but is not limited thereto. Then, a deep learning model can be used to classify each grid in the current grid map. For different application scenarios, each grid can be classified into different types. For example, in a mine work scenario, each grid can be classified as ground, static obstacle, boundary, etc., but is not limited thereto. Generally, the classification process of each grid usually involves processing the current grid map using a deep learning model to determine the probability of each grid belonging to each category, and then the category with the maximum probability can be determined as the category of the grid based on the probability.

[0072] Then the current grid map can be input into the grid map corresponding to the work area, and the recognition probability of each grid is accumulated and the height value is filled, that is, the historical probability of different grids in the historical map data can be updated based on the recognition probability of each grid in the current grid map, so as to obtain the target probability, for example, the result obtained by accumulating the probability that the corresponding grid belongs to each category in the historical map data can be updated based on the recognition probability of each grid in the current grid map. Optionally, the historical probability can be updated by using an empirical formula, for example, the forgetting rate can be determined based on experience, and then the target probability can be determined by using the following formula: target probability = forgetting rate * historical probability + recognition probability of the grid, but not limited to this, and can also be realized by a deep learning model and the like.

[0073] Finally, the boundary of the grid map can be extracted based on the target categories of different grids in the grid map corresponding to the work area by using a pattern recognition algorithm, a computational geometry algorithm and the like, so as to obtain the perception map boundary data.

[0074] It should be noted that the two-dimensional coordinate information of the center point has been recorded in each grid of the current grid map, and each grid can be filled with height information. The height information here can refer to the recognition probability of each grid, but is not limited to this.

[0075] Preferably, the current grid map is a result obtained by point cloud registration of the perception grid map and the grid map corresponding to the work area, and the perception grid map is a result obtained by accumulating a plurality of historical grid maps.

[0076] In an optional embodiment, in order to improve the accuracy of the grid map maintained locally by the second vehicle, the current grid map can be input into the grid map corresponding to the work area, and the current grid map is spliced and optimized. The splicing and optimization here can be point cloud registration of the current grid map and the grid map corresponding to the work area, and alignment of the locally maintained grid map.

[0077] Embodiment 2

[0078] According to the embodiments of the present disclosure, a vehicle control device is also provided, which can execute the vehicle control method in Embodiment 1 described above, and the specific implementation scheme and preferred application scenarios are the same as those of the above-mentioned embodiment, which will not be repeated here.

[0079] FIG. 5 is a schematic diagram of a vehicle control device according to an embodiment of the present disclosure, as shown in FIG. 5, the device comprises:

[0080] The acquisition module 52 is configured to acquire first perception information of the first vehicle in response to receiving the first spatial information of the first work site, wherein the first spatial information is generated based on the initial map boundary data, and the first perception information and the initial map boundary data are both used to determine the boundary of the work area.

[0081] The adjustment module 54 is configured to adjust the first spatial information based on the first perception information to obtain second spatial information of the first work site, wherein the matching degree between the second spatial information and the boundary of the work area is greater than the matching degree between the first spatial information and the boundary of the work area.

[0082] The generation module 56 is configured to generate a first control instruction based on the second spatial information, wherein the first control instruction is used to control the first vehicle to travel to the first work site.

[0083] In the above embodiments of the present disclosure, the adjustment module 54 includes: a determination unit configured to determine, based on the first spatial information and the first perception information, a first intrusion degree of the boundary of the work area corresponding to the first perception information with respect to the first work site corresponding to the first spatial information; and an adjustment unit configured to, in response to the first intrusion degree being less than a preset threshold, adjust the first spatial information to obtain the second spatial information, so that a second intrusion degree of the boundary of the work area corresponding to the first perception information with respect to the first work site corresponding to the second spatial information is greater than or equal to the preset threshold.

[0084] In the above embodiments of the present disclosure, the first spatial information at least includes: first position information and a first orientation of the first work site; the adjustment unit is configured to adjust the first position information according to the first orientation to obtain second position information; and the second spatial information is obtained based on the first orientation and the second position information.

[0085] In the above embodiments of the present disclosure, the generation module is configured to generate a second control instruction based on third spatial information of a to-be-worked site in response to receiving the third spatial information, wherein the to-be-worked site is used to park the first vehicle in the case that the first vehicle is parked at the first work site, the second control instruction is used to control the first vehicle to travel to the to-be-worked site; and the acquisition module is configured to acquire the first spatial information in response to the first vehicle traveling to the to-be-worked site.

[0086] In the above embodiments of the present disclosure, the first spatial information is spatial information generated based on a second orientation of a work device and third position information, the work device is used to work together with the vehicle at the first work site, and the third position information is determined based on initial map data sent by a cloud server, wherein the initial map data includes the initial map boundary data.

[0087] In the foregoing embodiments of the present disclosure, the initial map data is obtained by updating the historical map data of the work area based on the perception map boundary data sent by the second vehicle, wherein the perception map boundary data is the map boundary data generated by the second vehicle based on the second perception information of the second vehicle after the second vehicle drives out of the second work position.

[0088] In the foregoing embodiments of the present disclosure, the initial map data is obtained by replacing a set of boundary points between historical end points in the historical map data based on the perception map boundary data, wherein the historical end points are end points searched in a search tree corresponding to the historical map data based on the perception end points in the perception map boundary data, and the distance between the historical end points and the perception end points is less than the distance between other boundary points in the search tree and the perception end points, and the other boundary points are used to represent boundary points other than the historical end points in the search tree.

[0089] In the foregoing embodiments of the present disclosure, the perception map boundary data is obtained by boundary extraction on a grid map corresponding to the work area based on target categories of different grids in the grid map, wherein the target categories are determined based on target probabilities of the grids corresponding to multiple categories, the target probabilities are probabilities obtained by updating historical probabilities of the grids using recognition probabilities of the grids, the recognition probabilities are probabilities of the grids corresponding to the multiple categories obtained by recognizing a current grid map using a deep learning model, the current grid map is a result of grid processing on the second perception information, the grid map corresponding to the work area is a result of accumulation on multiple historical grid maps, and the historical probabilities are accumulation results of the probabilities of the grids corresponding to the multiple categories obtained by recognizing the multiple historical grid maps using the deep learning model.

[0090] Preferably, the current grid map is a result of point cloud registration on a perception grid map and the grid map corresponding to the work area, and the perception grid map is a result of accumulation on the multiple historical grid maps.

[0091] Embodiment 3

[0092] The embodiments of the present application also provide an electronic device, including a memory storing an executable program, and a processor configured to run the program, wherein the program performs the method in the embodiments of the present disclosure when running.

[0093] Embodiment 4

[0094] The embodiments of the present application also provide a computer-readable storage medium, which includes a stored executable program, wherein the executable program controls a device where the computer-readable storage medium is located to perform the method in the embodiments of the present disclosure when running.

[0095] Embodiment 5

[0096] Embodiments of the present application also provide a computer program product comprising a computer program which, when executed by a processor, implements the method in any of the embodiments of the present disclosure.

[0097] Embodiment 6

[0098] Embodiments of the present application also provide a computer program product comprising a non-volatile computer-readable storage medium arranged to store a computer program which, when executed by a processor, implements the method in any of the embodiments of the present disclosure.

[0099] Embodiment 7

[0100] Embodiments of the present application also provide a computer program which, when executed by a processor, implements the method in any of the embodiments of the present disclosure.

[0101] The above-mentioned sequence numbers of the embodiments of the present disclosure are for description only, and do not represent the advantages or disadvantages of the embodiments.

[0102] In the above-mentioned embodiments of the present disclosure, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0103] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other means. Among them, the above-mentioned device embodiments are illustrative, for example, the division of the units can be a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0104] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0105] In addition, each functional unit in each embodiment of the present disclosure can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0106] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present disclosure, essentially or in part, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods according to the embodiments of the present disclosure. The aforementioned storage medium includes: various memories (for example, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and the like) and the like that can store program codes.

[0107] The above describes the preferred embodiments of the present disclosure. It should be noted that, for those skilled in the art, without departing from the principles of the present disclosure, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present disclosure. Industrial applicability

[0108] The embodiments of the present disclosure provide a vehicle control method and device. In response to receiving first space information of a first work site, first perception information of a first vehicle is obtained, the first space information is adjusted based on the first perception information to obtain second space information of the first work site, and the first control instruction is generated based on the second space information to control the first vehicle to travel to the first work site. It is easy to note that after the first space information of the first work site is generated based on the initial map boundary data, the first space information can be adjusted through real-time perception of the first vehicle to obtain second space information with higher precision, and the first vehicle is controlled to travel based on the second space information, so that the first vehicle is parked in place, thereby realizing the generation of high-precision loading sites, improving the control precision of the unmanned mine truck, and further improving the technical effect of the work efficiency, thereby solving the technical problem that in the related art, map updating is affected by the network, resulting in reduced control precision of the unmanned mine truck and affecting the work efficiency.

Claims

1. A vehicle control method, comprising: obtaining first perception information of a first vehicle in response to receiving first spatial information of a first work site, wherein the first spatial information is generated based on initial map boundary data, and the first perception information and the initial map boundary data are both used to determine a boundary of a work area; adjusting the first spatial information based on the first perception information to obtain second spatial information of the first work site, wherein a matching degree between the second spatial information and the boundary of the work area is greater than a matching degree between the first spatial information and the boundary of the work area; generating a first control instruction based on the second spatial information, wherein the first control instruction is used to control the first vehicle to travel to the first work site.

2. The method of claim 1, wherein, The adjusting of the first spatial information based on the first perception information to obtain the second spatial information of the first work site comprises: determining a first intrusion degree of the boundary of the work area corresponding to the first perception information relative to the first work site corresponding to the first spatial information based on the first spatial information and the first perception information; in response to the first intrusion degree being less than a preset threshold, adjusting the first spatial information to obtain the second spatial information, so that a second intrusion degree of the boundary of the work area corresponding to the first perception information relative to the first work site corresponding to the second spatial information is greater than or equal to the preset threshold.

3. The method of claim 2, wherein, The first spatial information at least includes first position information and a first orientation of the first work site, and the adjusting of the first spatial information to obtain the second spatial information comprises: adjusting the first position information according to the first orientation to obtain second position information; and obtaining the second spatial information based on the first orientation and the second position information.

4. The method of claim 1, wherein, The method further comprises: generating a second control instruction based on third spatial information of a to-be-worked site in response to receiving the third spatial information, wherein the to-be-worked site is used to park the first vehicle in a case where a vehicle is parked on the first work site, and the second control instruction is used to control the first vehicle to travel to the to-be-worked site; obtaining the first spatial information in response to the first vehicle traveling to the to-be-worked site.

5. The method of claim 1, wherein, The first spatial information is spatial information generated based on a second orientation of a work device and third position information, the work device is used to work with a vehicle on the first work site, and the third position information is determined based on initial map data sent by a cloud server, wherein the initial map data includes the initial map boundary data.

6. The method of claim 1 or 5, wherein, The initial map data is obtained by updating historical map data of the work area based on perception map boundary data sent by a second vehicle, wherein the perception map boundary data is map boundary data generated based on second perception information of the second vehicle after the second vehicle travels out of a second work site.

7. The method of claim 6, wherein, The initial map data is data obtained by replacing a set of boundary points between historical endpoints in the historical map data based on the perception map boundary data, the historical endpoints being endpoints searched in a search tree corresponding to the historical map data based on a perception endpoint in the perception map boundary data, wherein a distance between the historical endpoints and the perception endpoint is less than a distance between other boundary points in the search tree and the perception endpoint, and the other boundary points are used to represent boundary points in the search tree other than the historical endpoints.

8. The method of claim 6, wherein, The perception map boundary data is data obtained by boundary extraction on a grid map corresponding to the work area based on target categories of different grids in the grid map, the target categories being determined based on a target probability of the grid corresponding to a plurality of categories, the target probability being a probability obtained by updating a historical probability of the grid using a recognition probability of the grid, and the recognition probability being a probability that the grid corresponds to the plurality of categories obtained by recognizing a current grid map using a deep learning model, the current grid map being a result of rasterizing the second perception information, and the grid map corresponding to the work area being a result of accumulation of a plurality of historical grid maps, and the historical probability being an accumulation result of the probability that the grid corresponds to the plurality of categories obtained by recognizing the plurality of historical grid maps using the deep learning model.

9. The method of claim 5, wherein, The second orientation and the third position information are determined based on operation instructions received by the work equipment, or determined based on third perception information of the work equipment, or determined based on work requirements of the work equipment.

10. A vehicle control apparatus, comprising: an acquisition module configured to acquire first perception information of a first vehicle in response to receiving first space information of a first work site, wherein the first space information is generated based on initial map boundary data, and the first perception information and the initial map boundary data are both used to determine boundaries of a work area; an adjustment module configured to adjust the first space information based on the first perception information to obtain second space information of the first work site, wherein a matching degree between the second space information and the boundaries of the work area is greater than a matching degree between the first space information and the boundaries of the work area; a generation module configured to generate a first control instruction based on the second space information, wherein the first control instruction is used to control the first vehicle to travel to the first work site.

Images