A robot control method and a robot

Abstract

The present application discloses a robot control method and a robot. Based on referring to the search history record, the robot control method further improves the efficiency and accuracy of the robot in finding a predetermined target by means of the information of the photometric map and the height map. At the same time, it can also help the robot find a better position that is not easily found by the user among many positions, thereby improving the intelligence of the robot and the user experience.

Description

Technical Field

[0001] The present application relates to the field of intelligent robots, and particularly to a robot control method and a robot. Background Art

[0002] Nowadays, home service robots represented by cleaning robots are becoming more and more popular. Many families choose to buy cleaning robots to help with housework, thereby improving the quality of life. However, cleaning robots are usually only used for cleaning the home environment and have relatively single functions. If some practical functions can be added additionally, it will greatly increase the fun of users using cleaning robots and at the same time increase the sense of happiness in life. For example, the invention application with the patent publication number CN108789432A discloses a hide-and-seek robot control system based on positioning logic. The robot of this application can play hide-and-seek with users through the hiding control system, increasing the entertainment items for children indoors at home and making children in a good mood. However, this application only realizes the hide-and-seek function through simple positioning and analysis, and the user experience is not good. Summary of the Invention

[0003] The present application provides a robot control method and a robot, and the specific technical solutions are as follows:

[0004] A robot control method, the method specifically includes the following steps: Step S1, based on the global map and the search history record, the robot selects a position that meets the first preset condition for target search. When all targets are searched, the search stops. When not all targets are searched at the position that meets the first preset condition, enter Step S2; Step S2, based on the photometric map and / or the height map, the robot selects a position that meets the second preset condition for target search. When all targets are searched, or the number of search times exceeds the preset number of times, or the search time exceeds the preset time, the search stops.

[0005] Further, in the Step S1, the method for the robot to select a position that meets the first preset condition for target search specifically includes: Step S11, the robot reads the search history record and selects several positions with the most times of finding targets; Step S12, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for targets; wherein, the search history record at least includes the coordinate information of the position on the global map and the historical number of times the robot finds targets at each position.

[0006] Further, in step S2, the method for the robot to select a position meeting the second preset condition for target searching specifically includes: step S21a, the robot reads the photometric map, and then uses a preset convolutional kernel to take the derivative of the photometric map to obtain a number of extreme points; step S22a, the robot clusters the number of extreme points and compares the numerical magnitudes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to a corresponding position on the photometric map, and the positions on the photometric map and the global map are in one-to-one correspondence; step S23a, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

[0007] Further, in step S2, the method for the robot to select a position meeting the second preset condition for target searching specifically includes: step S21b, the robot reads the height map, and then uses a preset convolutional kernel to take the derivative of the height map to obtain a number of extreme points; step S22b, the robot clusters the number of extreme points and compares the numerical magnitudes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to a corresponding position on the height map, and the positions on the height map and the global map are in one-to-one correspondence; step S23b, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

[0008] Further, in step S2, the method for the robot to select a position meeting the second preset condition for target searching specifically includes: step S21c, the robot reads the photometric map and the height map simultaneously, and then performs weighted summation on the photometric values and height values at the corresponding positions on the photometric map and the height map to obtain a fused map of the photometric map and the height map; step S22c, the robot uses a preset convolutional kernel to take the derivative of the fused map to obtain a number of extreme points; step S23c, the robot clusters the number of extreme points and compares the numerical magnitudes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to a corresponding position on the fused map, and the positions on the fused map and the global map are in one-to-one correspondence; step S24c, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

[0009] Further, the method for the robot to perform target searching specifically includes: after the robot moves to a position meeting the first preset condition or a position meeting the second preset condition, it looks around and uses a camera for target recognition. If the target is recognized, it moves to the next position or stops searching. If the target is not recognized, it recognizes a preset object, then makes a preset operation based on the preset object, and at the same time performs target recognition, and finally moves to the next position or stops searching.

[0010] A robot control method, the method specifically includes the following steps: Step Q1, the robot reads the photometric map, and then multiplies the photometric information on the photometric map by the first equalization coefficient respectively to obtain a number of first products; Step Q2, the robot reads the height map, and then multiplies the height information on the height map by the second equalization coefficient respectively to obtain a number of second products; Step Q3, the robot sums the first product and the second product at the corresponding positions of the photometric map and the height map to obtain the reference score for each position; wherein, the positions on the photometric map and the height map correspond one by one; Step Q4, the robot compares the reference scores of each position, and then selects the position with the largest reference score as the optimal position and moves to that position to perform the next operation.

[0011] Further, Step Q3 further includes the following steps: Step Q31, based on the global map, the robot calculates the path length between its current position and each position on the global map, and then multiplies them by the third equalization coefficient respectively to obtain a number of third products; Step Q32, the robot sums the first product, the second product and the third product at the corresponding positions of the photometric map, the height map and the global map to obtain the reference score for each position; wherein, the positions on the photometric map, the height map and the global map correspond one by one.

[0012] Further, Step Q32 further includes the following steps: Step Q321, the robot reads the search history record, and then multiplies the historical number of times the robot finds the target at each position on the global map by the fourth equalization coefficient respectively to obtain a number of fourth products; wherein, the search history record at least includes the coordinate information of the position on the global map and the historical number of times the robot finds the target at each position; Step Q322, the robot sums the first product, the second product, the third product and the fourth product at the corresponding positions of the photometric map, the height map and the global map to obtain the reference score for each position; wherein, the positions on the photometric map, the height map and the global map correspond one by one.

[0013] Further, Step Q322 further includes: after the robot sums the first product, the second product, the third product and the fourth product at the corresponding positions of the photometric map, the height map and the global map, it multiplies the sum by a random coefficient to obtain the reference score for each position.

[0014] A robot, the robot is used to implement the described robot control method, the robot includes: a camera, used to construct a global map and identify targets; a photometric sensor, used to obtain photometric information to generate a photometric map; a height sensor, used to obtain height information to generate a height map.

[0015] Further, the height sensor is a TOF sensor, and the TOF sensor is used to detect the distance to an obstacle above the robot.

[0016] The robot control method described in this application, on the basis of referring to the search history record, further improves the efficiency and accuracy of the robot to find the predetermined target through the information of the photometric map and the height map. At the same time, it can also help the robot find a better position that is not easily found by the user among many positions, thereby improving the intelligence of the robot and the user experience. Description of the Drawings

[0017] Figure 1 It is a schematic flowchart of the robot control method according to an embodiment of this application.

[0018] Figure 2 It is a schematic flowchart of the robot control method according to another embodiment of this application. Detailed Embodiments

[0019] In the following description, specific details such as specific system structures and technologies are proposed for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of this application. However, those skilled in the art should clearly understand that this application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of this application.

[0020] It should be understood that when used in this application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations. It should also be understood that the term "and / or" as used in this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0021] As used in this application, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" according to the context.

[0022] In addition, in the description of the present application, terms such as "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance. The reference to "an embodiment" or "some embodiments" etc. described in the specification of the present application means that specific features, structures or characteristics described in combination with the embodiment are included in one or more embodiments of the present application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments" etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. Terms such as "comprise", "include", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0023] The invention application with the patent publication number CN108789432A discloses a hide-and-seek robot control system based on positioning logic. The robot in this application can play hide-and-seek with users through the hiding control system, increasing the entertainment items for children in the indoor home environment and making children in a happy mood. However, this application only realizes the hide-and-seek function through simple positioning and analysis, resulting in a poor user experience.

[0024] To solve the above technical problems, the embodiment of the present application provides a robot control method. Based on the reference of the search history record, the efficiency and accuracy of the robot to find a predetermined target are further improved through the information of the photometric map and the height map. As Figure 1 shown, the method specifically includes the following steps: Step S1, based on the global map and the search history record, the robot selects a position that meets the first preset condition for target search. When all targets are found, the search stops. When not all targets are found at the position that meets the first preset condition, it enters Step S2; Step S2, based on the photometric map and / or the height map, the robot selects a position that meets the second preset condition for target search. When all targets are found, or the number of searches exceeds the preset number of times, or the search time exceeds the preset time, the search stops.

[0025] It should be noted that the global map is a map established by the robot using a mapping algorithm after collecting environmental information through environmental detection sensors such as cameras or lidar, which can reflect the overall working environment of the robot. The global map is a grid map. In this application, the size of one grid is taken as the size of one robot body. It can be understood by those skilled in the art that the grid map constructed immediately is marked with the environmental information around the current position of the robot. The grids within the map area constructed by the robot include three states: free, occupied, and unknown; these grids are represented by grid points in the motion trajectory line segment of this embodiment, that is, the center points of the grids; the grid points in the free state refer to the grids not occupied by obstacles, which are the grid position points that the robot can reach, and are free grid points, which can form an unoccupied area; the grid points in the occupied state refer to the grids occupied by obstacles, which are obstacle grid points and can form an occupied area; the unknown grid points refer to the position points where the specific situation is not clear during the process of the robot constructing the map, and are often blocked by obstacles, which can form an unknown area. In this application, the global map can completely cover the overall environment where the robot works.

[0026] As one of the implementation manners, in step S1, the method for the robot to select a position that meets the first preset condition for target search specifically includes: step S11, the robot reads the search history record and selects several positions with the most search times for the target; step S12, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target; where the search history record at least includes the coordinate information of the position on the global map and the historical number of times the robot has found the target at each position. In an embodiment, the robot searches for a user in a hide-and-seek game. The robot selects the top 5 positions with the most hiding times of the user from the search history record for inspection, which can improve the probability of finding the user and reflect the intelligence of the robot. Further, if the user is not found at the top 5 positions, 5 more positions can be randomly selected from the remaining positions for inspection. The 10 positions selected in this embodiment may be relatively scattered on the global map, involving a problem of the shortest total path. Therefore, the robot finds the solution with the shortest total path through the traveling salesman problem to improve the efficiency of finding people. When the user cannot be found according to the search history record, proceed to the next step.

[0027] As one of the implementation manners, in step S2, the method for the robot to select a position meeting the second preset condition for target search specifically includes: step S21a, the robot reads the photometric map, and then uses a preset convolution kernel to take the derivative of the photometric map to obtain a number of extreme points; step S22a, the robot clusters the number of extreme points and compares the numerical magnitudes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to a corresponding position on the photometric map, and the positions on the photometric map and the global map are in one-to-one correspondence; step S23a, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

[0028] Among them, the robot obtains the photometric map of the overall environment through a photometric sensor. The size of the photometric map is comparable to the size of the current map, and it is also a grid map. The meaning of each grid is the light intensity at that position, and the size of the grid is taken as the size of one body. The representation of the light intensity at the machine end is to map the value read by the photometric sensor into the interval [0, 255]. 0 represents the darkest, and 255 represents the brightest. If the robot detects different light intensities at the same position at two different times, the photometric information at that position takes the average value.

[0029] During the execution of step S21a, the robot uses an N*N convolution kernel to take the derivative of the photometric image in the X direction, and then rotates the convolution kernel by 90 degrees and takes the derivative in the Y direction on the derivative map to obtain multiple extreme points. Generally, there will be a situation where multiple extreme points are very close to each other. At this time, a clustering algorithm is needed to aggregate the extreme points to divide the closer extreme points into the same set, and then use the center point of the same set to represent all the extreme points in the set, simplifying the search mechanism of the robot and improving the efficiency. Optionally, the MeanShift algorithm is used to cluster the extreme points. In one embodiment, when the robot is looking for a user in a hide-and-seek game, it is preferably considered to check the positions with weaker light intensity to improve the hit rate. Therefore, among the clustered extreme points, select the positions corresponding to several extreme points with smaller numerical values for inspection.

[0030] As another implementation manner, in the step S2, the method for the robot to select a position meeting the second preset condition for target searching specifically includes: step S21b, the robot reads the height map, and then uses a preset convolution kernel to take the derivative of the height map to obtain a number of extreme points; step S22b, the robot clusters the number of extreme points and compares the numerical magnitudes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to a corresponding position on the height map, and the positions on the height map and the global map are in one-to-one correspondence; step S23b, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

[0031] Wherein, the robot obtains the height map of the overall environment through a height sensor. The size of the height map is equivalent to the size of the current map, and it is also a grid map. The meaning of each grid is the distance to the obstacle above the robot, that is, how high there is an obstacle above the robot's head, and the size of the grid is taken as the size of one body. If the robot detects different height information at the same position at two different times, the height information at this position takes the average value.

[0032] During the execution of step S21b, the robot also uses an N*N convolution kernel to take the derivative of the height image to obtain a number of extreme points, and then searches for the target at the position with a lower height to improve the hit rate. This process is similar to step S21a and will not be elaborated here.

[0033] As yet another implementation manner, in the step S2, the method for the robot to select a position meeting the second preset condition for target searching specifically includes: step S21c, the robot reads the photometric map and the height map simultaneously, and then performs weighted summation on the photometric values and height values at the corresponding positions on the photometric map and the height map to obtain a fused map of the photometric map and the height map; step S22c, the robot uses a preset convolution kernel to take the derivative of the fused map to obtain a number of extreme points; step S23c, the robot clusters the number of extreme points and compares the numerical magnitudes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to a corresponding position on the fused map, and the positions on the fused map and the global map are in one-to-one correspondence; step S24c, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

[0034] When the robot has both a photometric map and a height map, the data from both can be combined to select a more appropriate search location. First, the two maps are fused. Assume that L(i) is the i-th point of the photometric map, H(i) is the i-th point of the height map, and S(i) is the score of the photometric map and the height map at the i-th point. Calculate S(i) = a×L(i) + b×H(i), where a and b are the adjustment coefficients of the two maps, both of which are empirical values, and S(i) is the fused map of the photometric map and the height map. Then, the robot selects a location with weaker light intensity and lower height on the fused map to search for the target.

[0035] As one of the embodiments, the method for the robot to perform target search specifically includes: after the robot moves to a position that meets the first preset condition or a position that meets the second preset condition, it looks around and uses a camera to identify the target. If the target is identified, it moves to the next position or stops searching. If the target cannot be identified, it identifies a preset object, and then performs a preset operation based on the preset object while performing target identification, and finally moves to the next position or stops searching.

[0036] It should be noted that the robot pre-stores a model for identifying targets, and the model uses the YOLO framework to train targets (such as images containing user features). In one embodiment, the robot searches for users in a hide-and-seek game. There is a pre-collection phase before the game starts. The robot needs to collect information from players participating in the game separately, because there are other people in the environment besides the players. If the information is not collected, it will cause misjudgment and affect the game experience. If the player is sensitive to privacy issues such as biometric leakage, he can bring objects that can distinguish him from other players (such as headgear, etc.). The players collected in the pre-collection phase are all the objects of the robot's search.

[0037] When the robot does not recognize the target at a certain location, it will not stop searching immediately. The robot only calculates a location with a high probability where the player may hide, and the player usually hides under a covered object. Therefore, the robot will use the pre-trained model to identify the surrounding objects or furniture that are convenient for hiding, and then perform corresponding operations based on the recognized objects, such as turning on the lighting to improve the recognition rate or opening the door to search (with a robotic arm). When the robot recognizes the target at a certain location, it needs to confirm with the target. The confirmation process can be completed through voice interaction or through a smart terminal.

[0038] like Figure 2As shown in the figure, an embodiment of the present application provides a robot control method, which specifically includes the following steps: Step Q1, the robot reads the photometric map, and then multiplies the photometric information on the photometric map by the first equalization coefficient respectively to obtain a number of first products; Step Q2, the robot reads the height map, and then multiplies the height information on the height map by the second equalization coefficient respectively to obtain a number of second products; Step Q3, the robot sums the first product and the second product at the corresponding positions of the photometric map and the height map to obtain the reference score for each position; wherein, the positions on the photometric map and the height map correspond one by one; Step Q4, the robot compares the reference scores of each position, and then selects the position with the largest reference score as the optimal position and moves to this position to perform the next operation.

[0039] Compared with the robot searching for the user in the hide-and-seek game, the robot can only choose one position when hiding in the game. The strategy for choosing the hiding position is basically the same as the strategy for choosing the searching position, except that an optimal solution needs to be selected from many positions. The embodiment of the present application also helps the robot find a better position that is not easily found by the user from many positions based on the information provided by the photometric map and the height map, so as to improve the intelligence of the robot and the user experience.

[0040] As one of the implementation manners, Step Q3 further includes the following steps: Step Q31, based on the global map, the robot calculates the path length between its current position and each position on the global map, and then multiplies them by the third equalization coefficient respectively to obtain a number of third products; Step Q32, the robot sums the first product, the second product and the third product at the corresponding positions of the photometric map, the height map and the global map to obtain the reference score for each position; wherein, the positions on the photometric map, the height map and the global map correspond one by one. Considering the influence of the path length makes the robot's choice more reasonable and scientific.

[0041] As one of the implementation manners, step Q32 further includes the following steps: Step Q321, the robot reads the search history record, and then multiplies the historical number of times of finding the target at each position of the robot on the global map by the fourth balance coefficient respectively to obtain a number of fourth products; wherein, the search history record at least includes the coordinate information of the position on the global map and the historical number of times of finding the target by the robot at each position; Step Q322, the robot sums the first product, the second product, the third product and the fourth product at the corresponding positions of the photometric map, the height map and the global map to obtain the reference score of each position; wherein, the positions on the photometric map, the height map and the global map correspond one by one. Further considering the influence of the search history record makes the robot's selection more reasonable and scientific. For example, if position A is the place where the user has been found the most times in history, it indicates that position A is the key attention area of the user and should be avoided. It should be noted that the first balance coefficient, the second balance coefficient, the third balance coefficient and the fourth balance coefficient are all empirical values.

[0042] As one of the implementation manners, step Q322 further includes: after the robot sums the first product, the second product, the third product and the fourth product at the corresponding positions of the photometric map, the height map and the global map, it multiplies the sum by a random coefficient to obtain the reference score of each position. Adding randomness improves the fun of the game.

[0043] Based on the above implementation manner, the robot can select an optimal solution through the following formula:

[0044] S i = λ i (a i1 *LM i + a i2 *HM i + a i3 *D i + a i4 *H i )

[0045] Wherein, S i is the reference score of the i-th position, λ i is the random coefficient of the i-th position, a i1 is the first balance coefficient, a i2 is the second balance coefficient, a i3 is the third balance coefficient, a i4 is the fourth balance coefficient, LM i is the photometry of the i-th position, HM i is the height of the obstacle above the robot at the i-th position, D i is the path length from the i-th position to the starting position of the robot, H iThe historical number of times the robot has found the target at the i-th position.

[0046] When the robot is found in the hiding position, confirmation with the user is required. The user can indicate that they have found the robot through voice interaction, pressing a button, or a smart terminal. After receiving the information, the robot uses the camera to confirm the user. If it is the player, the game ends. If the robot's hiding time exceeds the limit, the game also ends, and the robot returns to the starting position or the base station.

[0047] 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, the references to memory, storage, database, or other media used in the various embodiments provided in this application can all include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory ROM, programmable memory PROM, electrically programmable memory DPROM, electrically erasable programmable memory DDPROM, or flash memory. Volatile memory can include random access memory RAM or external cache memory.

[0048] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations 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.

[0049] The above embodiments only represent several embodiments of this application. Their descriptions are relatively specific and detailed, but they should not be construed as a limitation on the scope of the patent application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application.

Claims

1. A robot control method, characterized in that, The method specifically includes the following steps: Step S1, based on the global map and the search history record, the robot selects a location that meets the first preset condition for target search. When all targets are found, the search stops. When not all targets are found at the location that meets the first preset condition, go to Step S2; Step S2, based on the photometric map and / or the height map, the robot selects a location that meets the second preset condition for target search. When all targets are found, or the number of searches exceeds the preset number, or the search time exceeds the preset time, the search stops; wherein, the photometric map is a grid map, and the photometric information on each grid represents the illumination intensity at that location. The height map is a grid map, and the height information on each grid represents the distance to the obstacle above when the robot is at that location; In the above Step S1, the method for the robot to select a location that meets the first preset condition for target search specifically includes: Step S11, the robot reads the search history record and selects several locations where the number of times of finding targets is the largest; Step S12, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each location to search for targets; Among them, the search history record at least includes the coordinate information of the location on the global map and the historical number of times the robot finds targets at each location; In the above Step S2, the method for the robot to select a location that meets the second preset condition for target search specifically includes: Step S21a, the robot reads the photometric map, and then uses a preset convolution kernel to take the derivative of the photometric map to obtain several extreme points; Step S22a, the robot clusters several extreme points and compares the numerical sizes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to the corresponding location on the photometric map, and the positions on the photometric map and the global map correspond one by one; Step S23a, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each location to search for targets; Or, Step S21b, the robot reads the height map, and then uses a preset convolution kernel to take the derivative of the height map to obtain several extreme points; Step S22b, the robot clusters several extreme points and compares the numerical sizes of the clustering results, and then selects a preset number of clustering results in ascending order; wherein, each clustering result corresponds to the corresponding location on the height map, and the positions on the height map and the global map correspond one by one; Step S23b, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each location to search for targets; Or, Step S21c, the robot reads the photometric map and the height map at the same time, and then performs weighted summation on the photometric values and height values at the corresponding positions on the photometric map and the height map to obtain a fused map of the photometric map and the height map; Step S22c, the robot uses a preset convolution kernel to take the derivative of the fused map to obtain several extreme points; Step S23c, the robot clusters a number of extreme points and compares the numerical values of the clustering results, and then selects a preset number of clustering results in ascending order; where each clustering result corresponds to a corresponding position on the fusion map, and the fusion map corresponds one-to-one with the positions on the global map; Step S24c, the robot finds the solution with the shortest total path through the traveling salesman problem, and then moves to each position to search for the target.

2. The robot control method according to claim 1, wherein The method for the robot to search for the target specifically includes: After the robot moves to a position that meets the first preset condition or a position that meets the second preset condition, it looks around and uses a camera for target recognition. If the target is recognized, it moves to the next position or stops searching. If the target is not recognized, it recognizes a preset object, then makes a preset operation based on the preset object, and at the same time performs target recognition, and finally moves to the next position or stops searching.

3. A robot control method, characterized in that, The method specifically includes the following steps: Step Q1, the robot reads the photometric map, and then multiplies the photometric information on the photometric map by the first equalization coefficient respectively to obtain a number of first products; Step Q2, the robot reads the height map, and then multiplies the height information on the height map by the second equalization coefficient respectively to obtain a number of second products; Step Q3, the robot sums the first product and the second product at the corresponding positions on the photometric map and the height map to obtain the reference score for each position; where the positions on the photometric map and the height map correspond one-to-one; Step Q4, the robot compares the reference scores of each position, and then selects the position with the largest reference score as the optimal position and moves to that position to perform the next operation; Among them, the photometric map is a grid map, and the photometric information on each grid represents the illumination intensity at that position. The height map is a grid map, and the height information on each grid represents the distance of the obstacle above the robot when it is at that position.

4. The robot control method according to claim 3, characterized in that, Step Q3 further includes the following steps: Step Q31, based on the global map, the robot calculates the path length between its current position and each position on the global map, and then multiplies them by the third equalization coefficient respectively to obtain a number of third products; Step Q32, the robot sums the first product, the second product and the third product at the corresponding positions on the photometric map, the height map and the global map to obtain the reference score for each position; where the positions on the photometric map, the height map and the global map correspond one-to-one.

5. A robot control method according to claim 4, characterized in that, Step Q32 further includes the following steps: Step Q321, the robot reads the search history record, and then multiplies the historical number of times the robot has found the target at each position on the global map by the fourth equalization coefficient respectively to obtain a number of fourth products; where the search history record at least includes the coordinate information of the position on the global map and the historical number of times the robot has found the target at each position; Step Q322, the robot sums the first product, the second product, the third product and the fourth product at the corresponding positions on the photometric map, the height map and the global map to obtain the reference score for each position; where the positions on the photometric map, the height map and the global map correspond one-to-one.

6. A robot control method according to claim 5, characterized in that, Step Q322 further includes: After the robot sums up the first product, the second product, the third product, and the fourth product at the corresponding positions of the photometric map, the height map, and the global map, it multiplies the sum by a random coefficient to obtain the reference score at each position.

7. A robot, characterized in that, The robot is used to implement the robot control method described in any one of claims 1 to 2 or claims 3 to 6. The robot includes: A camera, which is used to construct a global map and identify targets; A photometric sensor, which is used to obtain photometric information to generate a photometric map; A height sensor, which is used to obtain height information to generate a height map.

8. A robot according to claim 7, characterized in that, The height sensor is a TOF sensor, and the TOF sensor is used to detect the distance to the obstacle above the robot.

Images