Obstacle avoidance method of robot with body, robot with body, computer equipment and medium

Abstract

The invention relates to an obstacle avoidance method of a robot with a body, the robot with the body, computer equipment and a medium. Comprising the steps of determining obstacle associated data corresponding to a candidate position point for any candidate position point in a target area according to the number of times that the robot encounters a target obstacle at the candidate position point in a historical working process under the condition that the robot has a path planning demand; the target area is a non-first-time working area of the robot with the body; the obstacle association data is used for representing the probability that the target obstacle exists at the corresponding candidate position point; in a pre-constructed map corresponding to the target area, no obstacle exists under the candidate position point; determining a target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point; and controlling the robot with the body to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area. By adopting the method, the obstacle avoidance effect of the robot with the body can be improved.

Description

Technical Field

[0001] The present application relates to the field of embodied intelligent technology, and in particular to an obstacle avoidance method for an embodied robot, an embodied robot, a computer device and a medium. Background Art

[0002] With the rapid development of science and technology, people's living standards have been continuously improved, and embodied robots (for example, cleaning robots) have been widely used in people's daily lives.

[0003] Although embodied robots are widely used in daily life, their sensor technology still has limitations and it is difficult to fully predict all obstacles. Generally speaking, embodied robots cannot perform well in avoiding obstacles when facing blind spots, limited viewing angles, or dynamic environments (for example, opening and closing doors and curtains).

[0004] Based on this, there is an urgent need to provide an obstacle avoidance method to improve the obstacle avoidance effect and operation flexibility of the embodied robot. Summary of the invention

[0005] Based on this, it is necessary to provide an embodied robot obstacle avoidance method, embodied robot, computer device and medium that can improve the obstacle avoidance effect of the embodied robot in response to the above-mentioned technical problems.

[0006] In a first aspect, the present application provides an obstacle avoidance method for an embodied robot, comprising:

[0007] When the embodied robot has a need for path planning, for any candidate position point in the target area, the obstacle association data corresponding to the candidate position point is determined according to the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point;

[0008] Determine the target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point;

[0009] According to the target obstacle identifiers corresponding to the candidate position points in the target area, the embodied robot is controlled to avoid the target obstacles;

[0010] Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0011] In one embodiment, the number of times the embodied robot encounters a target obstacle at a candidate location during its historical work is determined based on the following method:

[0012] Obtain the latest triggering status of the embodied robot's environmental perception sensors at the candidate locations;

[0013] When the latest triggering situation represents that the environment perception sensor is triggered at the candidate position point, the number of times is increased by 1;

[0014] When the latest triggering situation indicates that the environmental perception sensor is not triggered at the candidate position point, the number of times is reduced by a preset value; the preset value is not less than 1; the minimum value of the number of times is 0, and the initial value of the number of times is 0.

[0015] In one embodiment, determining a target obstacle identifier corresponding to the candidate position point according to obstacle association data corresponding to the candidate position point includes:

[0016] The target obstacle identifier corresponding to the candidate position point is determined according to the size relationship between the obstacle association data corresponding to the candidate position point and the sample data corresponding to the candidate position point; and the value of the sample data is within a preset range.

[0017] In one of the embodiments, the preset interval is [0, 1]; the sample data is randomly sampled from a uniform distribution in the preset interval.

[0018] In one embodiment, determining the target obstacle identifier corresponding to the candidate position point according to the size relationship between the obstacle association data corresponding to the candidate position point and the sample data corresponding to the candidate position point includes:

[0019] When the obstacle-related data corresponding to the candidate position point is not greater than the sample data corresponding to the candidate position point, determining the target obstacle identifier corresponding to the candidate position point as the first identifier;

[0020] When the obstacle-related data corresponding to the candidate position point is greater than the sample data corresponding to the candidate position point, determining the target obstacle identifier corresponding to the candidate position point as the second identifier;

[0021] The first mark indicates that there is no target obstacle at the candidate location point; the second mark indicates that there is a target obstacle at the candidate location point.

[0022] In one embodiment, according to the target obstacle identifier corresponding to each candidate position point in the target area, controlling the embodied robot to avoid the target obstacle includes:

[0023] According to the target obstacle identification corresponding to each candidate position point in the target area, the initial obstacle identification corresponding to the corresponding candidate position point in the pre-constructed map of the target area is adjusted to generate a target map; wherein the target map is used to guide the embodied robot's current working process for the target area;

[0024] Based on the target map, a target planning path is generated; the target planning path is used to instruct the embodied robot to avoid target obstacles.

[0025] In one embodiment, according to the target obstacle identifiers corresponding to the candidate position points in the target area, adjusting the initial obstacle identifiers corresponding to the corresponding candidate position points in the pre-built map of the target area includes:

[0026] For any candidate position point, when the target obstacle identifier corresponding to the candidate position point is different from the corresponding initial obstacle identifier, the initial obstacle identifier corresponding to the corresponding candidate position point in the pre-constructed map of the target area is replaced with the target obstacle identifier.

[0027] In one embodiment, generating a target planning path based on a target map includes:

[0028] Based on the target map, the artificial potential field method is used to generate the target planning path.

[0029] In a second aspect, the present application further provides an embodied robot, comprising:

[0030] The first determination module is used to determine the obstacle association data corresponding to any candidate position point in the target area according to the number of times the embodied robot encounters the target obstacle at the candidate position point during the historical working process when there is a need for path planning for the embodied robot; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point;

[0031] A second determination module is used to determine the target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point;

[0032] The obstacle avoidance module is used to control the embodied robot to avoid the target obstacles according to the target obstacle identifiers corresponding to the candidate position points in the target area;

[0033] Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0034] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0035] When the embodied robot has a need for path planning, for any candidate position point in the target area, the obstacle association data corresponding to the candidate position point is determined according to the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point;

[0036] Determine the target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point;

[0037] According to the target obstacle identifiers corresponding to the candidate position points in the target area, the embodied robot is controlled to avoid the target obstacles;

[0038] Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0039] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0040] When the embodied robot has a need for path planning, for any candidate position point in the target area, the obstacle association data corresponding to the candidate position point is determined according to the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point;

[0041] Determine the target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point;

[0042] According to the target obstacle identifiers corresponding to the candidate position points in the target area, the embodied robot is controlled to avoid the target obstacles;

[0043] Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0044] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0045] When the embodied robot has a need for path planning, for any candidate position point in the target area, the obstacle association data corresponding to the candidate position point is determined according to the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point;

[0046] Determine the target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point;

[0047] According to the target obstacle identifiers corresponding to the candidate position points in the target area, the embodied robot is controlled to avoid the target obstacles;

[0048] Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0049] The obstacle avoidance method, embodied robot, computer device and medium of the embodied robot are as follows: before the embodied robot performs path planning for the target area, for any candidate position point in the target area, according to the number of times the embodied robot encounters the target obstacle at the candidate position point in the historical working process, the obstacle association data corresponding to the candidate position point is determined; and according to the obstacle association data corresponding to the candidate position point, the target obstacle identification corresponding to the candidate position point is determined; according to the target obstacle identification corresponding to each candidate position point in the target area, the embodied robot is controlled to avoid the target obstacle. Since the obstacle association data is the ratio between the number of times the embodied robot encounters the target obstacle at the candidate position point in the historical working process and the sum of the number and 1, it can characterize the probability of the target obstacle existing at the candidate position point to a certain extent. As the probability increases with the number of times the target obstacle is encountered, the obstacle association data will not be greatly affected by the embodied robot misjudging the candidate position point once or twice as encountering the target obstacle, thereby making the target obstacle identification corresponding to each candidate position point more reasonable and accurate. Furthermore, in the process of controlling the embodied robot to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area, a better obstacle avoidance effect can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 is a diagram of an application environment of an obstacle avoidance method for an embodied robot in one embodiment;

[0052] Figure 2 is a schematic flow chart of an obstacle avoidance method of an embodied robot in one embodiment;

[0053] Figure 3 A schematic flow chart of a step of determining the number of times a target obstacle is encountered in one embodiment;

[0054] Figure 4 A schematic diagram of a flow chart of a target obstacle identification determination step in one embodiment;

[0055] Figure 5 It is a flowchart of the steps of controlling the embodied robot to avoid a target obstacle in one embodiment;

[0056] Figure 6is a flow chart of an obstacle avoidance method of an embodied robot in another embodiment;

[0057] Figure 7 is a structural block diagram of an embodied robot in one embodiment;

[0058] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with 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.

[0060] The obstacle avoidance method of the embodied robot provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the embodied robot 102 communicates with the server 104 through the network. The embodied robot 102 includes but is not limited to a humanoid robot, a robotic arm, a cleaning robot (such as a sweeping robot, a mopping robot, or a sweeping and mopping robot), etc. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 can be an independent physical server, or it can be a server cluster or distributed system composed of multiple physical servers, or it can be a cloud server that provides cloud computing services.

[0061] In an exemplary embodiment, Figure 2 As shown, a method for avoiding obstacles of an embodied robot is provided, and the method is applied to Figure 1 The embodied robot in the example is used to illustrate, including the following steps:

[0062] S210, when the embodied robot has a need for path planning, for any candidate position point in the target area, the obstacle association data corresponding to the candidate position point is determined according to the number of times the embodied robot encounters a target obstacle at the candidate position point during its historical working process.

[0063] The target area may be a place where the embodied robot is about to perform various operations and tasks, that is, the work area to be performed by the embodied robot. Generally, the target area may be a non-first work area of ​​the embodied robot, that is, the embodied robot has completed work in the target area at some moments before the current moment. Taking a cleaning robot as an example, the target area may be an area that the cleaning robot is about to clean, and the cleaning robot has completed cleaning of the area before the current moment.

[0064] It can be understood that, usually, before the embodied robot performs a work task on the target area, the environmental data of the target area will be collected through the detection sensors on the embodied robot (such as lidar, visual sensors, etc.), so as to build a map corresponding to the target area based on the environmental data, that is, a pre-constructed map. In the pre-constructed map, the location points where fixed obstacles exist in the target area will be marked according to the environmental data. Among them, the position of fixed obstacles in the target area is relatively fixed and will not move at will, such as walls and cabinets. In the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point (that is, when the pre-constructed map corresponding to the target area is constructed based on the environmental data, the detection sensor does not detect an obstacle under the candidate position point).

[0065] Among them, for any candidate location point, the obstacle association data is used to characterize the probability that the target obstacle exists at the candidate location point. The target obstacle is a non-fixed obstacle, which can also be called a dynamic obstacle (such as a door that can be opened and closed, a curtain, etc.) or an undetectable obstacle. It can be understood that the target obstacle does not specifically refer to obstacles on the ground, but can also include obstacles in space, such as obstacles that the body or arms of the embodied robot may encounter; undetectable obstacles include transparent objects that the detection sensor cannot technically detect (such as a glass wall), or obstacles that cannot be detected due to the position limitation of the detection sensor installed on the embodied robot.

[0066] It can be understood that the path planning requirement can be generated by the embodied robot in response to the work instruction. Accordingly, in an optional embodiment, the embodied robot can respond to the work instruction to obtain the number of times each candidate position point encounters the target obstacle in the target area during the historical work of the embodied robot, and then determine the obstacle association data corresponding to the corresponding candidate position point according to each number. Exemplarily, for any candidate position point, the ratio of the number of times the embodied robot encounters the target obstacle at the candidate position point during the historical work of the embodied robot to the total number of times the embodied robot passes through the candidate position point during the historical work of the embodied robot can be used as the obstacle association data corresponding to the candidate position point.

[0067] In another optional implementation, for any candidate position point, the number of times the embodied robot encounters a target obstacle at the candidate position point during its historical work process can be input into a pre-trained association data determination model to obtain obstacle association data corresponding to the candidate position point. The association data determination model can be constructed based on a common neural network, which will not be described in detail here. Furthermore, in the process of model training the association data determination model, the number of samples can be input into the association data determination model to obtain predicted association data corresponding to the sample data, and then the model parameters of the association data determination model can be adjusted according to the predicted association data and the association data labels corresponding to the corresponding sample data to improve the model accuracy of the association data determination model.

[0068] In another optional embodiment, the correspondence between the reference number and the reference associated data can be determined in advance. Furthermore, for any candidate position point, the number of times the embodied robot encountered the target obstacle at the candidate position point during its historical working process and the reference associated data corresponding to the correspondence can be used as the obstacle associated data corresponding to the candidate position point.

[0069] In order to improve the referenceability and accuracy of obstacle association data, in another optional implementation, the obstacle association data corresponding to any candidate position point can be determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0070] Exemplarily, the obstacle association data corresponding to the candidate position point may be determined based on the following formula:

[0071] ;

[0072] In the formula, is the obstacle-related data corresponding to the position point (i, j) in the pre-built map; n is the number of times the embodied robot encounters the target obstacle at the position point (i, j) during its historical work; (i, j) is the coordinate position in the pre-built map.

[0073] It is worth noting that when n is 0, P ij is 0, indicating that there is no target obstacle at that location, which is consistent with the annotation on the pre-built map. ij When it approaches 1, it means there must be a target obstacle there. ij<1, which means that although the target obstacle is encountered here, it is not 100% certain that there is a target obstacle here, but only that the probability of the target obstacle existing here is high.

[0074] In addition, for the same candidate position point, the probability difference of the embodied robot encountering the target obstacle continuously at the candidate position point can be expressed as:

[0075] ;

[0076] In the formula, is the probability difference of the embodied robot encountering the target obstacle continuously at the candidate position; is the probability that the embodied robot encounters the target obstacle n times at the candidate position during its historical work process; P n-1 is the probability that the embodied robot encounters the target obstacle n-1 times at the candidate position during its historical work; n is the number of times the embodied robot encounters the target obstacle at position (i, j) during its historical work.

[0077] It can be seen that as n increases, That is, the impact of the embodied robot on the obstacle-related data will gradually decrease each time it encounters a target obstacle, rather than being consistent. Therefore, the embodied robot will not misjudge the candidate position point as encountering a target obstacle once or twice, which will have an impact on P. ij The obstacle association data determined based on the above formula has greater reference value and higher accuracy.

[0078] In addition, it is worth noting that in the limited historical work of the embodied robot, It will only approach 1, but will not be truly equal to 1. Therefore, the obstacle-related data will always have a certain probability of treating the candidate location point as a passable area and attempting to pass through it, which can avoid the problem of assuming that the candidate location point must have a target obstacle, so that in the subsequent execution of the task by the embodied robot, the target obstacle is always avoided, which improves the work efficiency of the embodied robot. However, as the number of obstacles encountered increases, the probability of attempting to pass through will become lower and lower.

[0079] S220: Determine a target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point.

[0080] The target obstacle identifier is used to characterize the existence of the target obstacle at the candidate location point.

[0081] In an optional implementation, a quantity threshold for determining the target obstacle identification can be pre-set, and the target obstacle identification corresponding to the candidate position point can be determined based on the size relationship between the obstacle association data and the quantity threshold. The quantity threshold can be determined based on manual experience, and this application does not make any limitation here. Exemplarily, the quantity threshold can be 0.5, and accordingly, when the obstacle association data is greater than 0.5, the target obstacle identification corresponding to the candidate position point is determined to be an obstacle; when the obstacle association data is not greater than 0.5, the target obstacle identification corresponding to the candidate position point is determined to be an obstacle that does not exist.

[0082] In another optional implementation, the obstacle association data can be input into a pre-trained identification determination model to obtain the target obstacle identification corresponding to the candidate position point. The identification determination model can be constructed based on a common neural network, which will not be described in detail here. Furthermore, in the process of training the identification determination model, the sample association data can be input into the identification determination model to obtain the predicted identification corresponding to the sample association data, and then the model parameters of the identification determination model can be adjusted according to the predicted identification and the identification label corresponding to the corresponding sample association data to improve the model accuracy of the identification determination model.

[0083] S230, controlling the embodied robot to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area.

[0084] Optionally, in the process of controlling the embodied robot to move in the target area, the embodied robot may be controlled to avoid candidate positions marked by the target obstacle as having obstacles, so as to control the embodied robot to avoid the target obstacle.

[0085] Typically, the path of the embodied robot for its work in the target area can be planned based on the target obstacle identifiers corresponding to the candidate position points in the target area, and the embodied robot can be controlled to work based on the planned path to avoid the target obstacles.

[0086] In the obstacle avoidance method of the embodied robot, before the embodied robot performs path planning for the target area, for any candidate position point in the target area, according to the number of times the embodied robot encounters the target obstacle at the candidate position point in the historical working process, the obstacle association data corresponding to the candidate position point is determined; and according to the obstacle association data corresponding to the candidate position point, the target obstacle identification corresponding to the candidate position point is determined; according to the target obstacle identification corresponding to each candidate position point in the target area, the embodied robot is controlled to avoid the target obstacle. Since the obstacle association data is the ratio between the number of times the embodied robot encounters the target obstacle at the candidate position point in the historical working process and the sum of the number and 1, it can characterize the probability of the target obstacle existing at the candidate position point to a certain extent. As the number of times the target obstacle is encountered increases, the obstacle association data will not be greatly affected by the embodied robot misjudging the target obstacle once or twice at the candidate position point, thereby making the target obstacle identification corresponding to each candidate position point more accurate. Further, in the process of controlling the embodied robot to avoid the target obstacle according to the target obstacle identification corresponding to each candidate position point in the target area, a better obstacle avoidance effect can be achieved.

[0087] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the process of determining the number of times the above embodied robot encounters a target obstacle at a candidate position during its historical working process is introduced.

[0088] For any candidate location point, see Figure 3 The step of determining the number of times the embodied robot encounters a target obstacle at the candidate position during the historical working process shown includes:

[0089] S310, obtaining the latest triggering status of the environment perception sensor of the embodied robot at the candidate position point.

[0090] Among them, the environmental perception sensor can help the embodied robot perceive obstacles, so that it can intelligently plan paths and avoid obstacles to adapt to different environments. The environmental perception sensor can include a collision sensor, specifically at least one of a contact sensor and a vibration sensor (such as a collision switch, an inertial measurement unit (IMU), a stress sensor, etc.) and a detection sensor.

[0091] Among them, the latest trigger situation represents the situation in which the embodied robot touches the target obstacle at the candidate position point during the most recent working process, which can include being triggered and not being triggered.

[0092] It is understandable that, during the movement of the embodied robot, if a collision is detected through the environmental perception sensor, but the pre-built map indicates that there is no obstacle at that location, it can be considered that the target obstacle has been encountered, that is, an undetectable obstacle or a dynamic obstacle.

[0093] Specifically, in this embodiment, the latest triggering status of the environment perception sensor at the candidate position point during the historical working process of the embodied robot can be obtained.

[0094] S320: When the latest triggering situation indicates that the environment perception sensor is triggered at the candidate location point, the number of times is increased by 1.

[0095] Among them, being triggered represents that the embodied robot touches the target obstacle at the candidate position during the most recent working process, that is, the environmental perception sensor detects a collision at the candidate position.

[0096] S330, when the latest triggering situation indicates that the environment perception sensor is not triggered at the candidate position point, the number of times is reduced by a preset value.

[0097] Among them, not triggered indicates that the embodied robot did not touch the target obstacle at the candidate position during the most recent work process, that is, the environmental perception sensor did not detect a collision at the candidate position.

[0098] Among them, the preset value is not less than 1; the minimum value of the number of times is 0, and the initial value of the number of times is 0.

[0099] Specifically, in this embodiment, for any candidate position point, the number of times the embodied robot touches the target obstacle at the candidate position point during its historical working process can be counted, and 1 is added each time an obstacle is encountered at the position, and 1 is subtracted when no obstacle is encountered, or a preset value of an integer greater than 1, such as 3 or 5, is used, until it is reduced to 0, or the current statistical number is directly subtracted (i.e., cleared to zero).

[0100] In the above embodiment, a specific method for determining the number of times the embodied robot encounters a target obstacle at each candidate position point during the historical working process is given, so that the determination of the number of times is clearer. Furthermore, for any candidate position point, when the latest trigger situation indicates that the environmental perception sensor is triggered at the candidate position point, the number is increased by 1; when the latest trigger situation indicates that the environmental perception sensor is not triggered at the candidate position point, the number is reduced by a preset value; the preset value is not less than 1, that is, when the latest trigger situation of the embodied robot indicates that the environmental perception sensor is not triggered at the candidate position point, it indicates that the embodied robot has not detected the target obstacle at the candidate position point this time, indicating that there is a high probability that there is no obstacle at the candidate position point. If the number of times the target obstacle is encountered before this time is not equal to 0, then the rate of reducing the number of times n of encountering the target obstacle is increased, which can make the obstacle association data determined based on the number more accurate.

[0101] Based on the technical solutions of the above embodiments, the present application further provides an optional embodiment. In this optional embodiment, the process of determining the target obstacle identification corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point is refined.

[0102] See also Figure 4 The target obstacle identification determination steps shown include:

[0103] S410, determining a target obstacle identifier corresponding to the candidate position point according to a size relationship between obstacle association data corresponding to the candidate position point and sample data corresponding to the candidate position point.

[0104] The value of the sample data is within a preset interval, which is [0, 1]. The sample data is randomly sampled from a uniform distribution in the preset interval, that is, the sample data is any random number between [0, 1], and the sample data corresponding to different obstacle-associated data may be different.

[0105] Exemplarily, when the obstacle-associated data corresponding to the candidate location point is not greater than the sample data corresponding to the candidate location point, the target obstacle identifier corresponding to the candidate location point is determined to be the first identifier; when the obstacle-associated data corresponding to the candidate location point is greater than the sample data corresponding to the candidate location point, the target obstacle identifier corresponding to the candidate location point is determined to be the second identifier.

[0106] The first mark indicates that there is no target obstacle at the candidate location point; the second mark indicates that there is a target obstacle at the candidate location point.

[0107] Exemplarily, the target obstacle identifier corresponding to the candidate position point may be determined based on the following formula:

[0108] ;

[0109] In the formula, is the target obstacle identifier corresponding to the candidate location point, When it is equal to 1, the corresponding target obstacle is identified as the second identification; When it is equal to 0, the corresponding target obstacle is identified as the first identification; is the obstacle association data corresponding to the position point (i, j) in the pre-built map; x is the sample data, , that is, x is a uniformly distributed sampling result between 0 and 1. When x< , it is considered that there is a target obstacle at the candidate position point and it should be avoided. At this time, the target obstacle mark corresponding to the candidate position point is the second mark; when x≥ , it is considered that there is no target obstacle at the candidate position point and normal passage is possible. At this time, the target obstacle identifier corresponding to the candidate position point is the first identifier.

[0110] For example, if the obstacle association data corresponding to a candidate position point is 0.5, and the corresponding sample data is 0.3, then the target obstacle identifier corresponding to the candidate position point is determined to be the second identifier, that is, it is determined that the target obstacle exists at the candidate position point. If the obstacle association data corresponding to a candidate position point is 0.5, and the corresponding sample data is 0.7, then the target obstacle identifier corresponding to the candidate position point is determined to be the first identifier, that is, it is determined that the target obstacle does not exist at the candidate position point.

[0111] It should be noted that x is a random number sampled uniformly between 0 and 1, which can be 0.1, 0.2, 0.3, etc. The probability of any random number sampled is the same, so when The larger the x< The greater the probability of , the greater the probability that there is a target obstacle at the candidate location point; The smaller, x< The smaller the probability of , the smaller the probability that the candidate location point has a target obstacle; Taking 0.8 as an example, x< The probability is 80%, that is, the probability that there is a target obstacle at the candidate location is 80%.

[0112] In the above embodiment, a specific method for determining the target obstacle identification corresponding to the candidate position point is provided. Specifically, based on the size relationship between the obstacle association data corresponding to the candidate position point and the corresponding sample data, the target obstacle corresponding to the corresponding candidate position point is determined, making the method for determining the target obstacle identification simpler. In addition, since the sample data is a random number between [0, 1], the probability of any random number sampled is the same. In the process of the robot executing a task a limited number of times, It will only approach 1, but will not be truly equal to 1. Therefore, through this sample data, there will always be a certain probability that the candidate location point will be treated as a passable area for attempt to pass through, but as the number of obstacles encountered increases, the probability of attempting to pass through will become lower and lower, making the target obstacle identification more referenceable.

[0113] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the step of controlling the embodied robot to avoid the target obstacle according to the target obstacle identification corresponding to each candidate position point in the target area is refined.

[0114] See also Figure 5 The steps of controlling the embodied robot to avoid a target obstacle include:

[0115] S510: According to the target obstacle identifiers corresponding to the candidate position points in the target area, the initial obstacle identifiers corresponding to the corresponding candidate position points in the pre-constructed map of the target area are adjusted to generate a target map.

[0116] Among them, for any candidate location point, its corresponding initial obstacle identifier is used to represent the obstacle identifier of the location point in the pre-built map. Usually, the initial obstacle identifier corresponding to the candidate location point is the first identifier, that is, there is no fixed obstacle at the candidate location point.

[0117] Among them, the target map is used to guide the embodied robot's current working process on the target area. That is to say, when the embodied robot works on the target area next time, its corresponding target map is still re-determined based on the method provided in the embodiment of the present application.

[0118] Exemplarily, for any candidate position point, when the target obstacle identifier corresponding to the candidate position point is different from the corresponding initial obstacle identifier, the initial obstacle identifier corresponding to the corresponding candidate position point in the pre-constructed map of the target area is replaced with the target obstacle identifier.

[0119] For example, the target obstacle identifiers corresponding to the candidate locations can be fused with the pre-built map of the target area to form a fused map, and the fused map is used as the target map. For example:

[0120] ;

[0121] In the formula, Characterize the target map; Represents the updated obstacle identification corresponding to the candidate location point (i, j) in the target map; is the target obstacle identifier corresponding to the candidate location point, When it is equal to 1, the corresponding target obstacle is identified as the second identification; When it is equal to 0, the corresponding target obstacle is identified as the first identification; Indicates the initial obstacle identifier corresponding to the candidate position point; It is the obstacle association data corresponding to the position point (i, j) in the pre-built map.

[0122] Based on this, the target obstacle identification corresponding to each candidate position point in the target map is updated.

[0123] S520, generating a target planning path based on the target map.

[0124] Among them, the target planning path is used to instruct the embodied robot to avoid target obstacles.

[0125] Exemplarily, based on the target map, an artificial potential field method can be used to generate a target planning path, so as to realize path planning by simulating the movement of the embodied robot in the attraction and repulsion force fields, thereby effectively avoiding obstacles; based on the target map, a grid method can be used to generate a target planning path; based on the target map, a random covering method can be used to generate a target planning path.

[0126] In the above embodiment, a specific method of controlling the embodied robot to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area is provided.

[0127] It is worth noting that during the process of the embodied robot working on the target area, it will record the latest triggering status of the embodied robot's environmental perception sensors at each candidate location point for next use.

[0128] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the obstacle avoidance method of the embodied robot provided by the present application is introduced in detail.

[0129] See also Figure 6 The obstacle avoidance method of the embodied robot shown comprises:

[0130] S610, determining the number of times each candidate position point in the target area encounters a target obstacle during the historical working process of the embodied robot;

[0131] S620, for any candidate position point, the number of times the embodied robot encounters a target obstacle at the candidate position point during its historical working process is used as first reference data;

[0132] S630, taking the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during the historical working process and 1 as the second reference data;

[0133] S640, taking the ratio of the first reference data to the second reference data as obstacle association data corresponding to the candidate position point;

[0134] S650, when the embodied robot has a path planning requirement, for any candidate position point in the target area, obtaining obstacle association data corresponding to the candidate position point;

[0135] S660A: When the obstacle-related data corresponding to the candidate position point is not greater than the sample data corresponding to the candidate position point, determine that the target obstacle identifier corresponding to the candidate position point is the first identifier;

[0136] S660B: when the obstacle-related data corresponding to the candidate position point is greater than the sample data corresponding to the candidate position point, determine that the target obstacle identifier corresponding to the candidate position point is the second identifier;

[0137] The first mark indicates that there is no target obstacle at the candidate location point; the second mark indicates that there is a target obstacle at the candidate location point.

[0138] S670: for any candidate position point, if the target obstacle identifier corresponding to the candidate position point is different from the corresponding initial obstacle identifier, replace the initial obstacle identifier corresponding to the corresponding candidate position point in the pre-constructed map of the target area with the target obstacle identifier;

[0139] Among them, the target map is used to guide the embodied robot's current working process in the target area;

[0140] S680, based on the target map, uses the artificial potential field method to generate the target planning path;

[0141] Among them, the target planning path is used to instruct the embodied robot to avoid target obstacles.

[0142] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, 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 can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0143] Based on the same inventive concept, the embodiment of the present application also provides an embodied robot for implementing the above-mentioned embodied robot obstacle avoidance method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodied robot embodiments provided below can refer to the limitations of the embodied robot obstacle avoidance method above, and will not be repeated here.

[0144] In an exemplary embodiment, Figure 7 As shown, an embodied robot is provided, including: a first determination module 710, a second determination module 720 and an obstacle avoidance module 730, wherein:

[0145] The first determination module 710 is used to determine the obstacle association data corresponding to any candidate position point in the target area according to the number of times the embodied robot encounters the target obstacle at the candidate position point during the historical working process when there is a need for path planning for the embodied robot; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point;

[0146] A second determination module 720 is used to determine a target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point;

[0147] The obstacle avoidance module 730 is used to control the embodied robot to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area;

[0148] Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

[0149] In one embodiment, the embodied robot also includes a third determination module, including an acquisition unit for acquiring the latest triggering situation of the embodied robot's environmental perception sensor at the candidate position point; a first update unit for adding 1 to the number of times when the latest triggering situation indicates that the environmental perception sensor is triggered at the candidate position point; a second update unit for subtracting a preset value from the number of times when the latest triggering situation indicates that the environmental perception sensor is not triggered at the candidate position point; the preset value is not less than 1; the minimum value of the number of times is 0, and the initial value of the number of times is 0.

[0150] In one embodiment, the second determination module 720 is specifically used to determine the target obstacle identifier corresponding to the candidate position point according to the size relationship between the obstacle association data corresponding to the candidate position point and the sample data corresponding to the candidate position point; the value of the sample data is within a preset interval. The preset interval is [0, 1]; the sample data is randomly sampled from a uniform distribution in the preset interval.

[0151] In one embodiment, the second determination module 720 includes a first determination unit, which is used to determine that the target obstacle identifier corresponding to the candidate location point is a first identifier when the obstacle association data corresponding to the candidate location point is not greater than the sample data corresponding to the candidate location point; a second determination unit, which is used to determine that the target obstacle identifier corresponding to the candidate location point is a second identifier when the obstacle association data corresponding to the candidate location point is greater than the sample data corresponding to the candidate location point; the first identifier indicates that there is no target obstacle at the candidate location point; and the second identifier indicates that there is a target obstacle at the candidate location point.

[0152] In one embodiment, the obstacle avoidance module 730 includes a first generation unit, which is used to adjust the initial obstacle identifications corresponding to the corresponding candidate position points in the pre-constructed map of the target area according to the target obstacle identifications corresponding to the candidate position points in the target area, so as to generate a target map; wherein the target map is used to guide the embodied robot's current working process for the target area; and a second generation unit is used to generate a target planning path based on the target map; the target planning path is used to instruct the embodied robot to avoid the target obstacles.

[0153] In one embodiment, the first generating unit is specifically used to replace the initial obstacle identifier corresponding to any candidate location point in the pre-constructed map of the target area with the target obstacle identifier when the target obstacle identifier corresponding to the candidate location point is different from the corresponding initial obstacle identifier.

[0154] In one embodiment, the second generating unit is specifically configured to generate a target planning path based on the target map by using an artificial potential field method.

[0155] Each module in the above-mentioned embodied robot can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0156] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device 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, near field communication (NFC) or other technologies. When the computer program is executed by the processor, an obstacle avoidance method of an embodied robot is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0157] Those skilled in the art will understand that Figure 8The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0158] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0159] 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 in the above method embodiments are implemented.

[0160] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0161] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and 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-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. 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), magnetic 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0162] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this application.

[0163] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, 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 attached claims.

Claims

1. An obstacle avoidance method for an embodied robot, characterized in that: The method comprises: In the case where the embodied robot has a need for path planning, for any candidate position point in the target area, the obstacle association data corresponding to the candidate position point is determined according to the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that the target obstacle exists at the corresponding candidate position point; in the pre-constructed map corresponding to the target area, there is no obstacle under the candidate position point; Determining a target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point; Controlling the embodied robot to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area; Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

2. The method according to claim 1, characterized in that: The number of times the embodied robot encounters a target obstacle at the candidate position during its historical operation is determined based on the following method: Obtain the latest triggering status of the environment perception sensor of the embodied robot at the candidate position point; When the latest triggering condition indicates that the environment perception sensor is triggered at the candidate position point, the number of times is increased by 1; When the latest triggering condition indicates that the environment perception sensor is not triggered at the candidate location point, the number of times is reduced by a preset value; the preset value is not less than 1; The minimum value of the number is 0, and the initial value of the number is 0.

3. The method according to claim 1, characterized in that: The determining, according to the obstacle association data corresponding to the candidate position point, a target obstacle identifier corresponding to the candidate position point comprises: The target obstacle identifier corresponding to the candidate position point is determined according to the size relationship between the obstacle association data corresponding to the candidate position point and the sample data corresponding to the candidate position point; and the value of the sample data is within a preset range.

4. The method according to claim 3, characterized in that The preset interval is [0, 1]; the sample data is randomly sampled from a uniform distribution in the preset interval.

5. The method according to claim 4, characterized in that The determining, according to the size relationship between the obstacle association data corresponding to the candidate position point and the sample data corresponding to the candidate position point, the target obstacle identifier corresponding to the candidate position point comprises: When the obstacle-related data corresponding to the candidate position point is not greater than the sample data corresponding to the candidate position point, determining that the target obstacle identifier corresponding to the candidate position point is the first identifier; When the obstacle-related data corresponding to the candidate position point is greater than the sample data corresponding to the candidate position point, determining that the target obstacle identifier corresponding to the candidate position point is a second identifier; The first mark indicates that there is no target obstacle at the candidate location point; the second mark indicates that there is a target obstacle at the candidate location point.

6. The method according to any one of claims 1 to 5, characterized in that The controlling the embodied robot to avoid the target obstacle according to the target obstacle identifier corresponding to each candidate position point in the target area includes: According to the target obstacle identifiers corresponding to the candidate positions in the target area, the initial obstacle identifiers corresponding to the corresponding candidate positions in the pre-constructed map of the target area are adjusted to generate a target map; wherein the target map is used to guide the embodied robot in the current working process of the target area; Based on the target map, a target planning path is generated; the target planning path is used to instruct the embodied robot to avoid target obstacles.

7. The method according to claim 6, characterized in that The adjusting, according to the target obstacle identifiers corresponding to the candidate position points in the target area, the initial obstacle identifiers corresponding to the corresponding candidate position points in the pre-built map of the target area comprises: For any candidate position point, when the target obstacle identifier corresponding to the candidate position point is different from the corresponding initial obstacle identifier, the initial obstacle identifier corresponding to the corresponding candidate position point in the pre-constructed map of the target area is replaced with the target obstacle identifier.

8. The method according to claim 6, characterized in that The generating a target planning path based on the target map includes: Based on the target map, an artificial potential field method is used to generate a target planning path.

9. An embodied robot, characterized in that: The embodied robot comprises: A first determination module is used to determine, for any candidate position point in a target area, obstacle association data corresponding to the candidate position point according to the number of times the embodied robot encounters a target obstacle at the candidate position point during its historical working process, when there is a need for path planning for the embodied robot; the target area is a non-first working area of ​​the embodied robot; the obstacle association data is used to characterize the probability that a target obstacle exists at the corresponding candidate position point; and in a pre-constructed map corresponding to the target area, no obstacle exists under the candidate position point; A second determination module, configured to determine a target obstacle identifier corresponding to the candidate position point according to the obstacle association data corresponding to the candidate position point; An obstacle avoidance module, used to control the embodied robot to avoid target obstacles according to target obstacle identifiers corresponding to each candidate position point in the target area; Among them, the obstacle association data corresponding to any candidate position point is determined based on the following method: the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process is used as the first reference data; the sum of the number of times the embodied robot encounters the target obstacle at the candidate position point during its historical working process and 1 is used as the second reference data; the ratio of the first reference data to the second reference data is used as the obstacle association data corresponding to the candidate position point.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images