Map updating method and device and robot

Abstract

The invention relates to a map updating method and device and a robot. The method comprises the steps of determining a target grid set covered by scanning in an initial environment grid map of a driving scene according to point cloud data obtained by scanning the driving scene by a robot, determining a target updating coefficient corresponding to each target grid according to the distance between each target grid in the target grid set and the robot, and updating the target updating coefficient according to the target updating coefficient. And updating the initial occupation state of each target grid in the initial environment grid map according to each target updating coefficient to obtain a target environment grid map. Wherein the target updating coefficient is negatively correlated with the distance. By means of the method, partition updating of the map can be achieved, appearance and disappearance of obstacles in a short-distance area in the map can be accelerated, appearance and disappearance of obstacles in a long-distance area in the map can be slowed down, the map updating efficiency is improved, and the timeliness that the updated map reflects real environment changes is enhanced.

Description

Technical Field

[0001] This application relates to the field of robot technology, and in particular, to a map updating method, apparatus, and robot. Background Art

[0002] During operation, a robot relies on a lidar carried by itself to sense and create a map of the surrounding environment and perform localization, that is, the Simultaneous Localization and Mapping (SLAM) technology.

[0003] In related technologies, to improve the accuracy of localization and mapping, it is necessary to update the environmental grid map in a timely manner. Among them, the main task of updating the environmental grid map is to update the occupancy status of the grids in the environmental grid map by obstacles, so as to achieve the update of the environmental grid map.

[0004] However, the map update efficiency in related technologies is low, resulting in poor timeliness in reflecting real environmental changes. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a map updating method, apparatus, and robot.

[0006] In a first aspect, this application provides a map updating method, including:

[0007] Determine a set of target grids covered by scanning in an initial environmental grid map of a driving scene according to point cloud data obtained by a robot scanning the driving scene;

[0008] Determine a target update coefficient corresponding to each target grid according to the distance between each target grid in the set of target grids and the robot; the target update coefficient is negatively correlated with the distance;

[0009] Update the initial occupancy status of each target grid in the initial environmental grid map according to each target update coefficient to obtain a target environmental grid map.

[0010] In one embodiment, determining a set of target grids covered by scanning in an initial environmental grid map of a driving scene according to point cloud data obtained by a robot scanning the driving scene includes:

[0011] Map the point cloud data to the initial environmental grid map to obtain scanning line segments between the robot and each point in the point cloud data in the initial environmental grid map;

[0012] Determine target grids intersecting the scanning line segments in the initial environmental grid map, and all the target grids form a set of target grids.

[0013] In one embodiment, the target update coefficients include an occupancy update coefficient corresponding to an occupied grid and a free update coefficient corresponding to a free grid; updating the initial occupancy status of each target grid in the initial environmental grid map according to the respective target update coefficients includes:

[0014] For each target grid, determine the status type of the target grid according to the positional relationship between the target grid and the points in the cloud data;

[0015] When the status type of the target grid is occupied, update the initial occupancy probability of the target grid according to the occupancy update coefficient;

[0016] When the status type of the target grid is free, update the initial free probability of the target grid according to the free update coefficient.

[0017] In one embodiment, updating the initial occupancy status of each target grid in the initial environmental grid map according to the respective target update coefficients includes:

[0018] Obtain the historical probability ratios of each target grid in the initial environmental grid map;

[0019] Update the initial occupancy status of each target grid according to the historical probability ratios and target update coefficients of each target grid.

[0020] In one embodiment, updating the initial occupancy status of each target grid according to the historical probability ratios and target update coefficients of each target grid includes:

[0021] For each target grid, determine the probability ratio of the target grid according to the historical probability ratio and the target update coefficient;

[0022] Determine the target occupancy status of each target grid according to the probability ratios of each target grid;

[0023] Replace the initial occupancy status with the target occupancy status of the same target grid to update the initial occupancy status of all target grids.

[0024] In one embodiment, determining the target update coefficient corresponding to each target grid according to the distance between each target grid in the target grid set and the robot includes:

[0025] For each target grid, obtain the distance between the target grid and the robot;

[0026] Determine the update coefficient corresponding to the distance between each target grid and the robot according to the preset correspondence between the distance and the update coefficient, and use it as the target update coefficient corresponding to each target grid.

[0027] In one embodiment, the process of constructing the correspondence between the distance and the update coefficient includes:

[0028] Obtain multiple groups of reference point cloud data scanned by the robot in different reference driving scenarios respectively; the reference driving scenarios include obstacles, and the preset distances between the obstacles and the robot in different reference driving scenarios are different;

[0029] For each reference driving scenario, determine the reference distance between the robot and the obstacle according to each group of reference point cloud data;

[0030] Construct the correspondence between the distance and the update coefficient according to each reference distance and the preset distance in each reference driving scenario.

[0031] In one embodiment, constructing the correspondence between the distance and the update coefficient according to each reference distance and the preset distance in each reference driving scenario includes:

[0032] For each reference driving scenario, obtain the distance deviation between each reference distance and the preset distance;

[0033] In all the reference distances of each reference driving scenario, count the first quantity of the reference distances whose distance deviations satisfy the deviation range, and the second quantity of the reference distances whose distance deviations do not satisfy the deviation range;

[0034] Construct the correspondence between the distance and the update coefficient according to the first quantity and the second quantity of each reference driving scenario.

[0035] In one embodiment, constructing the correspondence between the distance and the update coefficient according to the first quantity and the second quantity of each reference driving scenario includes:

[0036] For each reference driving scenario, obtain the ratio of the first quantity to the sum of the quantities as the occupancy update coefficient, and obtain the ratio of the second quantity to the sum of the quantities as the free update coefficient; the sum of the quantities is the sum of the first quantity and the second quantity;

[0037] Associate the preset distance and the corresponding occupancy update coefficient and free update coefficient of each reference driving scenario to obtain the correspondence between the distance and the update coefficient.

[0038] In a second aspect, the present application also provides a map update device, including:

[0039] A target determination module, configured to determine a set of target grids covered by scanning in the initial environmental grid map of the driving scenario according to the point cloud data scanned by the robot in the driving scenario;

[0040] A coefficient determination module, configured to determine a target update coefficient corresponding to each target grid according to the distance between each target grid in the target grid set and the robot; the target update coefficient is negatively correlated with the distance;

[0041] A state update module, configured to update the initial occupancy state of each target grid in the initial environmental grid map according to each target update coefficient to obtain a target environmental grid map.

[0042] In a third aspect, the present application further provides a robot, including a memory, a processor, a sensor, and a cleaning device. The memory stores a computer program, and when the processor executes the computer program, the steps of any one of the above methods are implemented.

[0043] In the above map update method, device, and robot, by obtaining point cloud data scanned by the robot for the driving scene, a target grid set covered by the scan is determined in the initial environmental grid map of the driving scene. According to the distance between each target grid in the target grid set and the robot, a target update coefficient corresponding to each target grid is determined, and the initial occupancy state of each target grid in the initial environmental grid map is updated according to each target update coefficient to obtain a target environmental grid map. In the above method, the target update coefficient is negatively correlated with the distance, and different distances correspond to different target update coefficients, so as to achieve zonal update of the map, which can accelerate the appearance and disappearance of obstacles in the short-distance area of the map and slow down the appearance and disappearance of obstacles in the long-distance area of the map, thereby reducing the impact of poor reliability of long-distance scanning on map update and reducing multiple scans for back-and-forth updates. Therefore, the map update efficiency is improved, and the timeliness of the updated map reflecting the real environmental changes is enhanced. Description of the Drawings

[0044] Figure 1 It is the internal structure diagram of the robot in one embodiment;

[0045] Figure 2 It is the flowchart of the map update method in one embodiment;

[0046] Figure 3 It is the flowchart of forming the target grid set in one embodiment;

[0047] Figure 4 It is a partial schematic diagram of the initial environmental grid map in one embodiment;

[0048] Figure 5 It is the flowchart of obtaining the target environmental grid map in one embodiment;

[0049] Figure 6 It is the flowchart of updating the initial occupancy state of the target grid in one embodiment;

[0050] Figure 7 Schematic diagram of the process for updating the initial occupancy state of the target grid in another embodiment;

[0051] Figure 8 Schematic diagram of the process for determining the target update coefficient corresponding to the target grid in one embodiment;

[0052] Figure 9 Schematic diagram of the process for constructing the correspondence between distance and update coefficient in one embodiment;

[0053] Figure 10 Schematic diagram of the process for constructing the correspondence between distance and update coefficient in another embodiment;

[0054] Figure 11 Schematic diagram of the process for constructing the correspondence between distance and update coefficient in another embodiment;

[0055] Figure 12 Schematic diagram of the process for the map update method in another embodiment;

[0056] Figure 13 Schematic diagram of the process for the map update method in another embodiment;

[0057] Figure 14 Structural block diagram of the map update device in one embodiment. Detailed implementation manners

[0058] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] The map update method provided by the embodiments of the present application can be applied to a robot, and its internal structural diagram can be as Figure 1As shown in the figure. The robot includes a processor, a memory, sensors, a cleaning device, a communication interface, a display unit, and an input device connected through a system bus. Among them, the processor of the robot is used to provide computing and control capabilities. The memory of the robot includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the robot is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a map update method. The display unit of the robot can be a liquid crystal display screen or an electronic ink display screen, and the input device of the robot can be a touch layer covering the display screen. The sensors of the robot are used to emit scanning signals to sense the driving environment. The sensors can be lidar. The cleaning device of the robot can include cleaning brushes.

[0060] Those skilled in the art can understand that Figure 1 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0061] In one embodiment, as Figure 2 shown, a map update method is provided. Taking the robot in Figure 1 as an example, the method includes the following steps:

[0062] S210. According to the point cloud data obtained by the robot scanning the driving scene, determine the set of target grids covered by the scan in the initial environmental grid map of the driving scene.

[0063] Among them, the set of target grids is the set of target grids covered by the scanning range of the robot in the initial environmental grid map.

[0064] Optionally, the robot can scan the driving scene through the lidar carried by itself, obtain the point cloud data collected by the lidar, and map the cloud data to the initial environmental grid map based on the positioning point of the robot, so as to determine the target grids according to the position distribution of the point cloud data in the initial environmental grid map, and thus obtain the set of target grids.

[0065] Exemplarily, the robot performs a 360° scan through lidar to obtain point cloud data, determines the grids occupied by the point cloud data in the initial environmental grid map, and sequentially connects the occupied grids to form a coverage area. All the grids within the coverage area are used as target grids to form a target grid set.

[0066] S220. Determine the target update coefficient corresponding to each target grid according to the distance between each target grid in the target grid set and the robot; the target update coefficient is positively correlated with the distance.

[0067] Among them, the target update coefficient is used to characterize the change rate of the occupancy state of the grid. The larger the target update coefficient, the greater the change rate of the occupancy state of the corresponding grid; conversely, the smaller the target update coefficient, the smaller the change rate of the occupancy state of the corresponding grid.

[0068] It should be noted that the scanning accuracy of the lidar varies with the distance of the object being measured. The closer the object being measured, the higher its accuracy; the farther the object being measured, the worse its accuracy. Based on this, the reliability of the point cloud data formed on the nearby obstacles in the scanned point cloud data is greater than that of the point cloud data formed on the distant obstacles. Therefore, a larger target update coefficient can be used to update the occupancy state of the target grids at close range, and a smaller target update coefficient can be used to update the occupancy state of the target grids at long range, that is, the target update coefficient is negatively correlated with the distance.

[0069] Optionally, after obtaining the target grid set, the robot can, for each target grid in the target grid set, obtain the distance between the position of the robot and the target grid, and determine the update coefficient corresponding to the distance, so as to obtain the target update coefficient corresponding to each target grid.

[0070] S230. Update the initial occupancy state of each target grid in the initial environmental grid map according to each target update coefficient to obtain a target environmental grid map.

[0071] Among them, the initial environmental grid map includes multiple grids, and each grid corresponds to an occupancy probability and an idle probability to characterize the occupancy state of the grid. The initial occupancy state of each grid in the initial environmental grid map represents the existing occupancy state of each grid before this update, and is cumulatively determined based on the historical point cloud data obtained from historical scans. In the case where this update is the first update, the initial environmental grid map is a blank grid map, and the occupancy probability of each grid is 0, which means it is not occupied.

[0072] It should be noted that the occupancy probability is used to characterize the possibility that the grid is occupied by an obstacle. The larger the occupancy probability, the greater the possibility that the corresponding grid is occupied by an obstacle; conversely, the smaller the occupancy probability, the smaller the possibility that the corresponding grid is occupied by an obstacle. The idle probability is used to characterize the possibility that the grid is idle and not occupied by an obstacle. The larger the idle probability, the smaller the possibility that the corresponding grid is not occupied by an obstacle; conversely, the smaller the idle probability, the greater the possibility that the corresponding grid is occupied by an obstacle. For the same grid, the sum of the occupancy probability and the idle probability is 1.

[0073] Optionally, for each target grid, the robot can use the target update coefficient corresponding to the target grid to update the initial occupancy state of the target grid in the initial environment grid map, obtain the updated occupancy state of the target grid, and implement the update of the initial occupancy state of all target grids to obtain the updated initial environment grid map, and use it as the target environment grid map obtained in this update.

[0074] In the embodiment of the present application, the target grid set covered by the scan is determined in the initial environment grid map of the driving scene according to the point cloud data obtained by the robot scanning the driving scene, the target update coefficient corresponding to each target grid is determined according to the distance between each target grid in the target grid set and the robot, and the initial occupancy state of each target grid in the initial environment grid map is updated according to each target update coefficient to obtain the target environment grid map. In the above method, different distances correspond to different target update coefficients to achieve the partition update of the map, and the target update coefficient is negatively correlated with the distance, that is, the closer the distance, the larger the target update coefficient, and the farther the distance, the smaller the target update coefficient, which can speed up the appearance and disappearance of obstacles in the short-distance area of ​​the map, and slow down the appearance and disappearance of obstacles in the long-distance area of ​​the map, thereby reducing the impact of the poor reliability of long-distance scanning on map updates, reducing multiple scanning back and forth updates, thereby improving the map update efficiency and enhancing the timeliness of the updated map reflecting real environmental changes.

[0075] To simplify the process of determining the target grid set, in one embodiment, as Figure 3 As shown, the above S210, based on the point cloud data obtained by the robot scanning the driving scene, determines the target grid set covered by the scan in the initial environment grid map of the driving scene, including:

[0076] S310, mapping the current cloud data to the initial environment grid map, and obtaining the scanning line segments between the robot in the initial environment grid map and each point in the point cloud data.

[0077] Optionally, after obtaining the point cloud data through scanning, the robot can map the point cloud data to the initial environmental grid map based on its own positioning point, and connect the position of the robot in the initial environmental grid map with each point in the point cloud data by straight lines to form the scanning line segments corresponding to each point. For example, as Figure 4 shown, the point cloud data obtained by the robot through scanning includes point A and point B. The robot uses point O as the position where the scanning signal is emitted, maps the point cloud data to the initial environmental grid map, and connects point O and point A with a straight line to form scanning line segment L1, and connects point O and point B with a straight line to form scanning line segment L2.

[0078] S320. Determine the target grids intersecting with the scanning line segments in the initial environmental grid map, and all the target grids form a target grid set.

[0079] Optionally, the robot can determine the target grids in the initial environmental grid map based on the positional relationship between all the scanning line segments formed by the point cloud data in the initial environmental grid map and the grids in the initial environmental grid map, and all the target grids form a target grid set. Among them, the robot can determine the intersection state of each scanning line segment and the grids in the initial environmental grid map, and determine the grids intersecting with the scanning line segments as the target grids. Continuing with the above example, Figure 4 in which grid 1 and grid 2 intersect with scanning line segment L1, and grids 3 - 10 intersect with scanning line segment L2. The robot can then determine that grids 1 - 10 in the initial environmental grid map are the target grids and form a target grid set {grids 1 - 10}.

[0080] In the embodiments of the present application, the point cloud data is mapped to the initial environmental grid map, and the scanning line segments between the robot in the initial environmental grid map and each point in the point cloud data are obtained to determine the target grids intersecting with the scanning line segments in the initial environmental grid map, and all the target grids form a target grid set. In the above method, the target grid set is determined based on the intersection state of the scanning line segments formed by the point cloud data and the grids in the initial environmental grid map, and the target grid set is determined by means of simple geometric relationships, which simplifies the process of determining the target grid set, improves the processing efficiency, and at the same time improves the reliability and accuracy of the obtained target grid set.

[0081] The target update coefficient includes the occupancy update coefficient corresponding to the occupied grid and the free update coefficient corresponding to the free grid. Based on this, in one embodiment, as Figure 5 shown, the update of the initial occupancy state of each target grid in the initial environmental grid map according to each target update coefficient in S230 above includes:

[0082] S510. For each target grid, determine the state type of the target grid according to the positional relationship between the target grid and the points in the point cloud data.

[0083] Among them, the occupied grid represents the occupied grid, and the free grid represents the free grid. The state type of the target grid is used to represent the real-time state type of whether the target grid determined based on the point cloud data is currently occupied or free.

[0084] Optionally, the robot can map the point cloud data to the initial environmental grid map. While determining the set of target grids, the state type of each target grid is determined based on the positional relationship between the target grid and the points in the point cloud data. Exemplarily, the robot can determine the state type of each target grid according to whether the target grid includes the points in the point cloud data, and record it as the occupancy state of each target grid. Among them, when the target grid includes the points in the point cloud data, the robot can determine that the state type of the target grid is occupied, and the target grid can be called a hit grid; when the target grid does not include the points in the point cloud data, the robot can determine that the state type of the target grid is free, and the target grid can be called a miss grid.

[0085] S520. When the state type of the target grid is occupied, update the initial occupancy probability of the target grid according to the occupancy update coefficient.

[0086] Optionally, after obtaining the state type of each target grid based on the point cloud data, for each target grid, when the state type of the target grid is occupied, the robot can obtain the occupancy update coefficient corresponding to the distance between the target grid and the robot, so as to update the initial occupancy probability of the target grid according to the occupancy update coefficient. Exemplarily, when the state type of the target grid is occupied, the robot can obtain the product of the occupancy update coefficient corresponding to the target grid and the existing initial occupancy probability of the target grid as the updated occupancy probability of the target grid.

[0087] S530. When the state type of the target grid is free, update the initial free probability of the target grid according to the free update coefficient.

[0088] Optionally, after obtaining the state type of each target grid based on the point cloud data, for each target grid, when the state type of the target grid is free, the robot can obtain the free update coefficient corresponding to the distance between the target grid and the robot, so as to update the initial free probability of the target grid according to the free update coefficient. Exemplarily, when the state type of the target grid is free, the robot can obtain the product of the free update coefficient corresponding to the target grid and the existing initial free probability of the target grid as the updated free probability of the target grid.

[0089] In the embodiments of the present application, the target update coefficient includes an occupancy update coefficient corresponding to an occupied grid and a free update coefficient corresponding to a free grid. Based on this, for each target grid, the state type of the target grid is determined according to the positional relationship between the target grid and the points in the point cloud data. When the state type of the target grid is occupied, the initial occupancy probability of the target grid is updated according to the occupancy update coefficient; when the state type of the target grid is free, the initial free probability of the target grid is updated according to the free update coefficient. In the above method, the state types of the target grids are determined based on the point cloud data, and different update coefficients are used for the target grids of different state types to update the initial occupancy state, so as to make the grid classification finer and improve the accuracy of map update.

[0090] In practical applications, map update is continuously carried out. Therefore, in one of the embodiments, as Figure 6 shown, updating the initial occupancy state of each target grid in the initial environmental grid map according to each target update coefficient in S230 above includes:

[0091] S610. Obtain the historical probability ratio of each target grid in the initial environmental grid map; the historical probability ratio of the target grid is the ratio of the occupancy probability to the free probability of the target grid after the previous update of the initial environmental grid map.

[0092] Among them, the probability ratio is the ratio of the occupancy probability to the free probability of the grid. The definition of the probability ratio odds is as follows:

[0093]

[0094]

[0095] s represents the grid state, s = 1 means the grid is occupied by an obstacle, and s = 0 means the grid is free. z represents the observation time of the grid scanned by the scanning signal, z = 1 means the hit event of scanning an obstacle, and z = 0 means the miss event of not scanning an obstacle.

[0096] Conditional probability represents the occupancy probability of the grid at the current t moment, which is the current occupancy probability of the grid obtained according to all the scanning observation information from historical time 1 to t (i.e., historical point cloud data); the conditional probability represents the free probability of the grid at the current t moment, which is the current free probability of the grid obtained according to all the scanning observation information from historical time 1 to t (i.e., historical point cloud data).

[0097] Optionally, the robot can directly obtain the probability ratio of each target grid in the initial environmental grid map as the historical probability ratio. For the current update of the initial environmental grid map, the historical probability ratio can be obtained based on the previous update of the initial environmental grid map. Among them, in the case where the current update is the first update, the historical probability ratio of all grids is 1.

[0098] S620. Update the initial occupancy status of each target grid according to the historical probability ratio of each target grid and the target update coefficient.

[0099] Optionally, after obtaining the historical probability ratio of each target grid, the robot can update the initial occupancy status of each target grid according to the correlation between the historical probability ratio and the target update coefficient.

[0100] It should be noted that according to the Markov property and by using the definition formula of the probability ratio in Bayesian localization for derivation, the correlation between the historical probability ratio and the update coefficient is as follows:

[0101]

[0102] Among them, t represents the current update corresponding to the initial environmental grid map, represents the target update coefficient, represents the historical probability ratio determined based on the historical point cloud data from time 1 to t - 1, corresponding to the previous update of the initial environmental grid map.

[0103] It should be noted that in the above correlation between the historical probability ratio and the update coefficient, that is, , has two values, namely the occupancy update coefficient and the free update coefficient.

[0104] The occupancy update coefficient actually represents the update coefficient value corresponding to the occurrence of the hit event , where p(s = 1|z = 1) represents the probability that there is actually an obstacle in the grid when an obstacle is observed in the grid during scanning.

[0105] The free update coefficient actually represents the update coefficient value corresponding to the occurrence of the miss event , where p(s = 1|z = 0) represents the probability that there is actually an obstacle in the grid when nothing is observed in the grid during scanning.

[0106] In an optional embodiment, as Figure 7 shown, the above S620. Update the initial occupancy status of each target grid according to the historical probability ratio of each target grid and the target update coefficient, includes:

[0107] S710. For each target grid, determine the probability ratio of the target grid according to the historical probability ratio and the target update coefficient.

[0108] Among them, the determined probability ratio of the target grid is the ratio of the occupancy probability to the free probability of the target grid after this update of the initial environmental grid map.

[0109] Optionally, after the robot obtains the correlation relationship between the historical probability ratio and the target update coefficient, based on this correlation relationship, the product of the historical probability ratio and the target update coefficient of the target grid can be obtained as the probability ratio of the target grid after this update. Among them, in the correlation relationship between the above historical probability ratio and the target update coefficient That is, it represents the probability ratio of the target grid determined based on the historical point cloud data from time 1 to time t - 1, corresponding to this update of the initial environmental grid map.

[0110] S720. Determine the target occupancy status of each target grid according to the probability ratio of each target grid.

[0111] Optionally, the robot can determine the occupancy status corresponding to the probability ratio of each target grid according to the conversion relationship between the probability ratio of the grid and the occupancy status, as the target occupancy status of the target grid.

[0112] It should be noted that the occupancy status of the grid can be characterized by the occupancy probability and / or the free probability, and is obtained based on the above definition formula of the probability ratio:

[0113] The conversion relationship between the probability ratio and the occupancy probability is as follows:

[0114]

[0115] The conversion relationship between the probability ratio and the free probability is as follows:

[0116]

[0117] Among them, when the target grid is an occupied grid, the occupancy status of the target grid is characterized by the occupancy probability; when the target grid is a free grid, the occupancy status of the target grid is characterized by the free probability.

[0118] Optionally, for each target grid, when the target grid is an occupied grid, the robot can determine the occupancy probability corresponding to the probability ratio of the target grid based on the conversion relationship between the probability ratio and the occupancy probability, as the target occupancy status of the target grid.

[0119] Among them, it can be known from the definition formula of the probability ratio that the occupancy probability corresponding to the probability ratio can be obtained based on the following formula:

[0120]

[0121] When the target grid is an idle grid, the robot can determine the idle probability corresponding to the probability ratio of the target grid based on the conversion relationship between the probability ratio and the idle probability, and use it as the target occupancy state of the target grid. In this way, the target occupancy states of all target grids are obtained.

[0122] S730. Replace the initial occupancy state with the target occupancy state of the same target grid to update the initial occupancy states of all target grids.

[0123] Optionally, after obtaining the target occupancy states of each target grid, the robot replaces the initial occupancy state with the target occupancy state for the same target grid to update the initial occupancy states of all target grids. Among them, for the same occupied grid among all target grids, the robot replaces the initial occupancy probability with the target occupancy probability of the occupied grid; for the same idle grid in the target grid, the robot replaces the initial idle probability with the target idle probability of the idle grid.

[0124] It should be noted that the update of the initial environmental grid map is essentially to update the occupancy probability of the grid. After obtaining the target idle probability of the idle grid, the robot can determine the target occupancy probability of the idle grid based on the fact that the sum of the occupancy probability and the idle probability of the same grid is 1, so as to update the occupancy probability of the target grid.

[0125] In the embodiment of the present application, the initial occupancy states of each target grid are updated by obtaining the historical probability ratio of the target grid in the initial environmental grid map, and according to the historical probability ratio and the target update coefficient of each target grid. Among them, the historical probability ratio of the target grid is the ratio of the occupancy probability to the idle probability of the target grid after the previous update of the initial environmental grid map. For each target grid, determine the probability ratio of the target grid according to the historical probability ratio and the target update coefficient, and determine the target occupancy state of each target grid according to the probability ratios of each target grid, so as to replace the initial occupancy state with the target occupancy state of the same target grid to update the initial occupancy states of all target grids. In the above method, the initial occupancy state of the target grid is updated based on the historical probability ratio and the target update coefficient. Each time an update is made, only the target update coefficient of the target grid corresponding to the point cloud data needs to be determined, and the historical probability ratio of the target grid obtained from the previous update can be directly obtained, which simplifies the map update process and improves the stability of the map update.

[0126] In practical applications, the update coefficient corresponding to each target grid can be determined based on the preset correspondence between the distance and the update coefficient. Based on this, in one embodiment, such as Figure 8As shown in the figure, in step S220, determining the target update coefficient corresponding to each target grid according to the distance between each target grid in the target grid set and the robot includes:

[0127] S810. For each target grid, obtain the distance between the target grid and the robot.

[0128] Optionally, after obtaining the target grid set, the robot can obtain the distance between the position of the robot and the center point of each target grid in the target grid set for each target grid in the target grid set, so as to obtain the distance corresponding to each target grid.

[0129] S820. Determine the update coefficient corresponding to the distance between each target grid and the robot according to the preset correspondence between the distance and the update coefficient, and use it as the target update coefficient corresponding to each target grid.

[0130] Wherein, each target grid corresponds to a distance and a target update coefficient at the same time. Exemplarily, the distance and the target update coefficient can be in one-to-one correspondence, such as one distance corresponding to one target update coefficient, or many-to-one, such as one distance range (including multiple distances) corresponding to one target update coefficient.

[0131] Optionally, after obtaining the distance corresponding to each target grid, the robot determines the update coefficient corresponding to the distance between each target grid and the robot according to the preset correspondence between the distance and the update coefficient, and uses it as the target update coefficient corresponding to each target grid.

[0132] In the embodiment of the present application, for each target grid, the distance between the target grid and the robot is obtained, and the update coefficient corresponding to the distance between each target grid and the robot is determined according to the preset correspondence between the distance and the update coefficient, and used as the target update coefficient corresponding to each target grid. In the above method, the correspondence between the distance and the update coefficient is pre-constructed, so that the target update coefficient corresponding to each target grid is determined by looking up the correspondence according to the distance between the target grid and the robot, which improves the efficiency and convenience of determining the target update coefficient, and correspondingly improves the map update efficiency and timeliness.

[0133] Next, the process of constructing the correspondence between the distance and the update coefficient will be introduced in detail. In one embodiment, as Figure 9 shown, the process of constructing the correspondence between the distance and the update coefficient includes:

[0134] S910. Obtain multiple groups of reference point cloud data scanned by the robot in different reference driving scenarios respectively; the reference driving scenarios include obstacles, and the preset distances between the obstacles and the robot in different reference driving scenarios are different.

[0135] Among them, the reference driving scenario is a robot driving scenario with obstacles pre-built.

[0136] Optionally, the robot can run in different reference driving scenarios respectively to obtain multiple groups of reference point cloud data scanned by the robot in different reference driving scenarios. Exemplarily, different reference driving scenarios include reference driving scenario A, reference driving scenario B, and reference driving scenario C. The preset distance between the obstacle and the robot in reference driving scenario A is LA, the preset distance between the obstacle and the robot in reference driving scenario B is LB, and the preset distance between the obstacle and the robot in reference driving scenario C is LC. The robot then performs multiple 360° scans in reference driving scenarios A, B, and C respectively. For example, if it performs 10 scans, it can obtain 10 groups of reference point cloud data scanned in reference driving scenario A, 10 groups of reference point cloud data scanned in reference driving scenario B, and 10 groups of reference point cloud data scanned in reference driving scenario C respectively.

[0137] S920. For each reference driving scenario, determine the reference distance between the robot and the obstacle according to each group of reference point cloud data.

[0138] Among them, each reference driving scenario corresponds to multiple groups of reference point cloud data and the same number of reference distances. Continuing the above example, each group of reference driving scenarios corresponds to 10 groups of reference point cloud data, and correspondingly 10 reference distances.

[0139] Optionally, for each reference driving scenario, the robot can respectively determine the point cloud data corresponding to the obstacle setting orientation in each group of reference point cloud data, so as to determine the distance between the robot and the obstacle in the corresponding reference driving scenario according to the point cloud data corresponding to the obstacle setting orientation, and use it as the reference distance, so as to obtain multiple reference distances corresponding to each reference driving scenario.

[0140] S930. Construct the corresponding relationship between the distance and the update coefficient according to each reference distance and the preset distance in each reference driving scenario.

[0141] Among them, the preset distance in each reference driving scenario is the actual distance between the robot and the obstacle, and the reference distance in each reference driving scenario is the measured distance between the robot and the obstacle.

[0142] Optionally, after the robot obtains each reference distance and the preset distance in each reference driving scenario, it can estimate the scanning accuracy of the robot corresponding to the preset distance based on each reference distance and the preset distance, so as to determine the corresponding update coefficient according to the estimated scanning accuracy, thereby associating the preset distance with the update coefficient and constructing the corresponding relationship between the distance and the update coefficient.

[0143] Among them, the higher the scanning accuracy, the correspondingly larger the update coefficient; the lower the scanning accuracy, the correspondingly smaller the update coefficient. Usually, the larger the preset distance, the lower the scanning accuracy; the smaller the preset distance, the lower the scanning accuracy.

[0144] In the embodiments of the present application, by respectively obtaining multiple groups of reference point cloud data scanned by the robot in different reference driving scenarios, for each reference driving scenario, the reference distance between the robot and the obstacle is determined according to each group of reference point cloud data, and then the corresponding relationship between the distance and the update coefficient is constructed according to each reference distance and the preset distance in each reference driving scenario. In the above method, the reference driving scenario includes obstacles, and the corresponding relationship between the distance and the update coefficient is constructed based on the reference distance between the robot and the obstacle measured by the robot in different reference driving scenarios and the preset distance actually set in the reference driving scenario, and the preset distance between the obstacle and the robot in different reference driving scenarios is different, which greatly improves the data richness of the corresponding relationship between the constructed distance and the update coefficient, so that each target grid can obtain the corresponding update coefficient, and correspondingly improves the success rate of map update.

[0145] The scanning accuracy can be characterized by the distance deviation between the reference distance and the preset distance in the same reference driving scenario. Based on this, in one embodiment, as Figure 10 shown, the above S930, constructing the corresponding relationship between the distance and the update coefficient according to each reference distance and the preset distance in each reference driving scenario, includes:

[0146] S1010. For each reference driving scenario, obtain the distance deviation between each reference distance and the preset distance.

[0147] Optionally, after obtaining multiple reference distances in each reference driving scenario, the robot can, for each reference driving scenario, obtain the difference between each reference distance and the preset distance as the distance deviation, so as to obtain the distance deviation corresponding to each reference distance in all reference driving scenarios.

[0148] Exemplarily, continuing the above example, for the reference driving scenario A, corresponding to the preset distance LA and 10 reference distances La, the robot obtains the difference between each reference distance La and the preset distance LA to obtain 10 distance deviations △A; for the reference driving scenario B, corresponding to the preset distance LB and 10 reference distances Lb, the robot obtains the difference between each reference distance Lb and the preset distance LB to obtain 10 distance deviations △B; similarly, for the reference driving scenario C, corresponding to the preset distance LC and 10 reference distances Lc, the robot obtains the difference between each reference distance Lc and the preset distance LC to obtain 10 distance deviations △C.

[0149] S1020. Among all the reference distances in each reference driving scenario, count the first quantity of the reference distances whose distance deviations meet the deviation range, and the second quantity of the reference distances whose distance deviations do not meet the deviation range.

[0150] Optionally, after obtaining the distance deviations corresponding to all the reference distances in each reference driving scenario, for each reference driving scenario, the robot can compare each distance deviation with a preset deviation range to determine whether the distance deviation meets the deviation range, and count the quantity of the reference distances whose distance deviations meet the deviation range among all the reference distances, denoted as the first quantity, and count the quantity of the reference distances whose distance deviations do not meet the deviation range among all the reference distances, denoted as the second quantity. Among them, if the distance deviation is within the preset deviation range, it is determined that the distance deviation meets the deviation range; conversely, if the distance deviation is not within the preset deviation range, it is determined that the distance deviation does not meet the deviation range.

[0151] Exemplarily, continuing the above example, among the 10 distance deviations △A corresponding to the reference driving scenario A, 7 meet the deviation range and 3 do not meet the deviation range, so the first quantity for the reference driving scenario A is 7 and the second quantity is 3; among the 10 distance deviations △B corresponding to the reference driving scenario B, 5 meet the deviation range and 5 do not meet the deviation range, so the first quantity for the reference driving scenario B is 5 and the second quantity is 5; among the 10 distance deviations △C corresponding to the reference driving scenario C, 4 meet the deviation range and 6 do not meet the deviation range, so the first quantity for the reference driving scenario C is 4 and the second quantity is 6.

[0152] S1030. Construct the corresponding relationship between the distance and the update coefficient according to the first quantity and the second quantity of each reference driving scenario.

[0153] Optionally, after obtaining the first quantity and the second quantity of each reference driving scenario, the robot can determine the update coefficient based on the first quantity and the second quantity, and associate the preset distance of the corresponding reference driving scenario with the update coefficient to obtain a set of corresponding relationships between the preset distance and the update coefficient. Multiple reference driving scenarios correspond to multiple preset distances, so multiple sets of corresponding relationships between the preset distance and the update coefficient can be obtained accordingly, so as to construct the corresponding relationship between the distance and the update coefficient based on all the corresponding relationships between the preset distance and the update coefficient.

[0154] Exemplarily, continuing with the above example, for the reference driving scenario A, the corresponding relationship T1 between the preset distance LA and the update coefficient Ca is obtained; for the reference driving scenario B, the corresponding relationship T2 between the preset distance LB and the update coefficient Cb is obtained; for the reference driving scenario C, the corresponding relationship T3 between the preset distance Lc and the update coefficient Cc is obtained; the corresponding relationships T1, T2, and T3 are summarized as the corresponding relationship between the distance and the update coefficient constructed.

[0155] In an optional embodiment, as Figure 11 shown, S1030 above, constructing the corresponding relationship between the distance and the update coefficient according to the first quantity and the second quantity of each reference driving scenario includes:

[0156] S1110. For each reference driving scenario, obtain the ratio of the first quantity to the sum of the quantities as the occupancy update coefficient, and obtain the ratio of the second quantity to the sum of the quantities as the idle update coefficient; the sum of the quantities is the sum of the first quantity and the second quantity.

[0157] Wherein, the sum of the quantities, that is, the sum of the first quantity and the second quantity, is the total number of scans of the robot in each reference driving scenario.

[0158] Optionally, for each reference driving scenario, the robot obtains the ratio of the statistically obtained first quantity to the sum of the quantities as the occupancy update coefficient, and obtains the ratio of the second quantity to the sum of the quantities as the idle update coefficient, so as to obtain the occupancy update coefficient and the idle update coefficient corresponding to all reference driving scenarios.

[0159] Exemplarily, continuing with the above example, for the reference driving scenario A, the occupancy update coefficient is 7 / 10, and the idle update coefficient is 3 / 10; for the reference driving scenario B, the occupancy update coefficient is 5 / 10, and the idle update coefficient is 5 / 10; for the reference driving scenario C, the occupancy update coefficient is 4 / 10, and the idle update coefficient is 6 / 10.

[0160] S1120. Associate the preset distance under each reference driving scenario with the corresponding occupancy update coefficient and idle update coefficient to obtain the corresponding relationship between the distance and the update coefficient.

[0161] Optionally, after obtaining the occupancy update coefficient and the idle update coefficient corresponding to all reference driving scenarios, the robot can associate the preset distance of each reference driving scenario with the corresponding occupancy update coefficient and idle update coefficient to obtain the association relationship between multiple preset distances and the occupancy update coefficient and the idle update coefficient, and group the preset distances with similar preset distances corresponding to the same set of occupancy update coefficient and idle update coefficient, so as to obtain the corresponding relationship between the distance and the update coefficient.

[0162] Exemplarily, continuing with the above example, for the reference driving scenario A, the corresponding relationship between the preset distance LA and the occupancy update coefficient of 0.7 and the free update coefficient of 0.3 is obtained; for the reference driving scenario B, the corresponding relationship between the preset distance LB and the occupancy update coefficient of 0.5 and the free update coefficient of 0.5 is obtained; for the reference driving scenario C, the corresponding relationship between the preset distance LC and the occupancy update coefficient of 0.4 and the free update coefficient of 0.6 is obtained. Among them, LA < LB < LC, and the corresponding relationship between the distance L and the update coefficient can be obtained as follows:

[0163] When L ≤ LA, the corresponding occupancy update coefficient is 0.7 and the free update coefficient is 0.3;

[0164] When LA < L ≤ LB, the corresponding occupancy update coefficient is 0.5 and the free update coefficient is 0.5;

[0165] When LB < L ≤ LC, the corresponding occupancy update coefficient is 0.4 and the free update coefficient is 0.6.

[0166] In the embodiments of the present application, for each reference distance in each reference driving scenario, the distance deviation between the reference distance and the preset distance is obtained. Among all the reference distances in each reference driving scenario, the first quantity of the reference distances whose distance deviation satisfies the deviation range and the second quantity of the reference distances whose distance deviation does not satisfy the deviation range are counted. According to the first quantity and the second quantity in each reference driving scenario, the corresponding relationship between the distance and the update coefficient is constructed. Among them, for the first quantity and the second quantity in each reference driving scenario, the ratio of the first quantity to the sum of the quantities is obtained as the occupancy update coefficient, and the ratio of the second quantity to the sum of the quantities is obtained as the free update coefficient. The preset distances and the corresponding occupancy update coefficients and free update coefficients in each reference driving scenario are associated to obtain the corresponding relationship between the distance and the update coefficient; the sum of the quantities is the sum of the first quantity and the second quantity. In the above method, the corresponding relationship between the distance and the update coefficient obtained includes different update coefficients corresponding to different distances. Using this corresponding relationship, the partition update of the map can be realized, thereby improving the map update efficiency and timeliness.

[0167] For the convenience of understanding by those skilled in the art, the map update method provided by the present application will be introduced in detail below, as Figure 12 shown, the method may include:

[0168] S1201. Map the point cloud data scanned by the robot for the driving scenario to the initial environmental grid map of the driving scenario, and obtain the scanning line segments between the robot and each point in the point cloud data in the initial environmental grid map;

[0169] S1202. Determine the target grids intersecting with the scanned line segment in the initial environmental grid map, and all the target grids form a target grid set;

[0170] S1203. For each target grid, obtain the distance between the target grid and the robot;

[0171] S1204. For each target grid, determine the state type of the target grid according to the positional relationship between the target grid and the points in the point cloud data;

[0172] S1205. When the state type of the target grid is occupied, determine the occupancy update coefficient corresponding to the distance between the target grid and the robot according to the corresponding relationship between the preset distance and the update coefficient, and update the initial occupancy probability of the target grid according to the occupancy update coefficient;

[0173] S1206. When the state type of the target grid is free, determine the free update coefficient corresponding to the distance between the target grid and the robot according to the corresponding relationship between the distance and the update coefficient, and update the initial free probability of the target grid according to the free update coefficient.

[0174] Among them, as Figure 13 shown, the process of constructing the corresponding relationship between the distance and the update coefficient includes:

[0175] S1301. Obtain multiple groups of reference point cloud data scanned by the robot in different reference driving scenarios respectively; the reference driving scenarios include obstacles, and the preset distances between the obstacles and the robot in different reference driving scenarios are different;

[0176] S1302. For each reference driving scenario, determine the reference distance between the robot and the obstacle according to each group of reference point cloud data;

[0177] S1303. For each reference driving scenario, obtain the distance deviation between each reference distance and the preset distance;

[0178] S1304. Among all the reference distances in each reference driving scenario, count the first quantity of the reference distances whose distance deviations satisfy the deviation range, and the second quantity of the reference distances whose distance deviations do not satisfy the deviation range;

[0179] S1305. For each reference driving scenario, obtain the ratio of the first quantity to the sum of the quantities as the occupancy update coefficient, and obtain the ratio of the second quantity to the sum of the quantities as the free update coefficient; the sum of the quantities is the sum of the first quantity and the second quantity;

[0180] S1306. Associate the preset distances of each reference driving scenario with the corresponding occupancy update factor and idle update factor to obtain the corresponding relationship between the distance and the update factor.

[0181] It should be noted that for the descriptions in S1201 - S1206 and S1301 - S1306 above, reference can be made to the relevant descriptions in the above embodiments, and their effects are similar. Therefore, they will not be elaborated in this embodiment.

[0182] It should be understood that although the steps in the flowcharts involved in the above - mentioned embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above - mentioned embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0183] Based on the same inventive concept, the embodiments of the present application also provide a map update device for implementing the above - mentioned map update method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the following map update device can refer to the limitations on the map update method in the above text, and will not be elaborated here.

[0184] In one embodiment, as Figure 14 shown, a map update device is provided, including: a target determination module 1401, a coefficient determination module 1402, and a state update module 1403, where:

[0185] The target determination module 1401 is configured to determine a set of target grids covered by scanning in the initial environmental grid map of the driving scenario according to the point cloud data scanned by the robot for the driving scenario;

[0186] The coefficient determination module 1402 is configured to determine the target update coefficient corresponding to each target grid according to the distance between each target grid in the set of target grids and the robot; the target update coefficient is negatively correlated with the distance;

[0187] The state update module 1403 is configured to update the initial occupancy state of each target grid in the initial environmental grid map according to each target update coefficient to obtain a target environmental grid map.

[0188] Each module in the above map updating device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in the processor of the computer device in hardware form or independent thereof, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0189] In one embodiment, a robot is provided, which includes a memory, a processor, a sensor, and a cleaning device. A computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the above methods are implemented.

[0190] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0191] In one embodiment, a computer program product is provided, which includes a computer program. When the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0192] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0193] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0194] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for map updating, characterized in that, The method includes: Determining a set of target grids covered by scanning in an initial environmental grid map of the driving scenario according to point cloud data obtained by the robot scanning the driving scenario; Determining a target update coefficient corresponding to each target grid according to the distance between each target grid in the set of target grids and the robot; the target update coefficient is negatively correlated with the distance; Updating the initial occupancy status of each target grid in the initial environmental grid map according to each target update coefficient to obtain a target environmental grid map.

2. The method according to claim 1, characterized in that, The determining a set of target grids covered by scanning in an initial environmental grid map of the driving scenario according to point cloud data obtained by the robot scanning the driving scenario includes: Mapping the point cloud data to the initial environmental grid map to obtain scanning line segments between the robot in the initial environmental grid map and each point in the point cloud data; Determining target grids intersecting with the scanning line segments in the initial environmental grid map, and all the target grids form the set of target grids.

3. The method according to claim 1, characterized in that, The target update coefficient includes an occupancy update coefficient corresponding to an occupied grid and an idle update coefficient corresponding to an idle grid; The updating the initial occupancy status of each target grid in the initial environmental grid map according to each target update coefficient includes: For each target grid, determining the state type of the target grid according to the positional relationship between the target grid and the points in the point cloud data; When the state type of the target grid is occupied, updating the initial occupancy probability of the target grid according to the occupancy update coefficient; When the state type of the target grid is idle, updating the initial idle probability of the target grid according to the idle update coefficient.

4. The method according to any one of claims 1 to 3, characterized in that, The updating the initial occupancy status of each target grid in the initial environmental grid map according to each target update coefficient includes: Obtaining the historical probability ratio of each target grid in the initial environmental grid map; the historical probability ratio of the target grid is the ratio of the occupancy probability to the idle probability of the target grid after the previous update of the initial environmental grid map; Updating the initial occupancy status of each target grid according to the historical probability ratio of each target grid and the target update coefficient.

5. The method according to claim 4, characterized in that The updating the initial occupancy status of each target grid according to the historical probability ratio of each target grid and the target update coefficient includes: For each target grid, determining the probability ratio of the target grid according to the historical probability ratio and the target update coefficient; Determining the target occupancy status of each target grid according to the probability ratio of each target grid; Replacing the initial occupancy status with the target occupancy status of the same target grid to update the initial occupancy status of all the target grids.

6. The method according to any one of claims 1 to 3, characterized in that, The determining a target update coefficient corresponding to each target grid according to the distance between each target grid in the set of target grids and the robot includes: For each target grid, obtaining the distance between the target grid and the robot; Determine the update coefficient corresponding to the distance between each of the target grids and the robot according to the corresponding relationship between the preset distance and the update coefficient, and use it as the target update coefficient corresponding to each of the target grids.

7. The method according to claim 6, characterized in that, The process of constructing the corresponding relationship between the distance and the update coefficient includes: Obtain multiple groups of reference point cloud data scanned by the robot in different reference driving scenarios respectively; the reference driving scenarios include obstacles, and the preset distances between the obstacles and the robot in different reference driving scenarios are different; For each of the reference driving scenarios, determine the reference distance between the robot and the obstacle according to each group of the reference point cloud data; Construct the corresponding relationship between the distance and the update coefficient according to each reference distance and the preset distance in each of the reference driving scenarios.

8. The method according to claim 7, characterized in that, The constructing the corresponding relationship between the distance and the update coefficient according to each reference distance and the preset distance in each of the reference driving scenarios includes: For each of the reference driving scenarios, obtain the distance deviation between each of the reference distances and the preset distance; In all the reference distances of each of the reference driving scenarios, count the first quantity of the reference distances whose distance deviations satisfy the deviation range, and the second quantity of the reference distances whose distance deviations do not satisfy the deviation range; Construct the corresponding relationship between the distance and the update coefficient according to the first quantity and the second quantity of each of the reference driving scenarios.

9. The method according to claim 8, wherein The constructing the corresponding relationship between the distance and the update coefficient according to the first quantity and the second quantity of each of the reference driving scenarios includes: For each of the reference driving scenarios, obtain the ratio of the first quantity to the sum of the quantities as the occupancy update coefficient, and obtain the ratio of the second quantity to the sum of the quantities as the idle update coefficient; the sum of the quantities is the sum of the first quantity and the second quantity; Associate the preset distance and the corresponding occupancy update coefficient and idle update coefficient of each of the reference driving scenarios to obtain the corresponding relationship between the distance and the update coefficient.

10. A map update device, characterized in that, The device includes: A target determination module, configured to determine a set of target grids covered by scanning in the initial environment grid map of the driving scenario according to the point cloud data scanned by the robot for the driving scenario; A coefficient determination module, configured to determine the target update coefficient corresponding to each of the target grids according to the distance between each of the target grids in the target grid set and the robot; the target update coefficient is negatively correlated with the distance; A state update module, configured to update the initial occupancy state of each of the target grids in the initial environment grid map according to each of the target update coefficients to obtain a target environment grid map.

11. A robot, comprising a memory, a processor, a sensor, and a cleaning device, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 9.

Images