Humanoid robot based on multi-cooperative intelligent algorithm and control method thereof

Abstract

The invention relates to the technical field of humanoid robots, in particular to a humanoid robot based on a multi-cooperative intelligent algorithm and a control method thereof, and the humanoid robot comprises a skeleton structure, a servo motor, a sensing module, a data acquisition unit, a first analysis unit, a second analysis unit, an adjustment unit and a correction unit, the sensing module is used for collecting real-time videos of the environment and obstacle position data, the data obtaining unit obtains the data, the first analysis unit constructs a sampling circle, calculates a risk degree evaluation value and guides the robot to move, and the second analysis unit corrects the preset length when the humanoid robot stops moving and does not arrive at the destination. The problems that in the prior art, due to the fact that a single sensor is adopted, obstacle information cannot be accurately reflected by surrounding environment monitoring data, meanwhile, an appropriate mode for adjusting an obstacle avoidance algorithm to cope with different obstacle environments is lacked, and the obstacle avoidance accuracy and safety of a humanoid robot are poor are solved.

Description

Technical Field

[0001] This invention relates to the field of humanoid robot technology, and in particular to a humanoid robot based on a multi-cooperative intelligent algorithm and its control method. Background Technology

[0002] In the development of robotics, humanoid robots have received widespread attention due to their resemblance to humans and intuitive operation. These robots are typically designed to perform complex tasks, such as serving, assisting, or performing specific tasks in human-inhabited environments. To effectively navigate and perform tasks in these environments, humanoid robots must possess a high degree of environmental awareness and intelligent decision-making capabilities.

[0003] Traditional humanoid robots face numerous challenges in navigation and obstacle avoidance. First, they need to be able to perceive and understand their surroundings in real time, including recognizing the location and size of obstacles. Second, robots need to be able to react quickly to environmental conditions and adjust their paths to avoid obstacles. Furthermore, as environmental complexity increases, robots also need to be able to handle more obstacles and more confined spaces.

[0004] Chinese Patent Application Publication No. CN117250959A discloses a dynamic multi-obstacle avoidance method for humanoid robots based on bidirectional A* and local perception. The method includes: S1 Overall path planning: Based on initial basic map information, a bidirectional A* search algorithm is used to determine the optimal route for the humanoid robot, and the direction of the path is used as an auxiliary route for the robot's movement; S2 Local obstacle avoidance: When the position the humanoid robot is about to move at coincides with the position of a dynamic obstacle, a local perception algorithm is executed, specifically including: S21: A sensor is placed in front of the humanoid robot's direction of travel to perceive the environment ahead, and a tracking matrix is ​​added to update dynamic obstacle information in a timely manner; S22: Combining the information provided by the tracking matrix, it is determined whether a feasible region exists in front of the humanoid robot; S23: If a feasible region exists, the humanoid robot is guided to choose a feasible direction to move forward; if no feasible region exists, the humanoid robot returns to the previous position, updates the dynamic obstacle information in real time, and executes the auxiliary route; S3: Combining overall path planning and local obstacle avoidance, the path planning of the humanoid robot in a dynamic multi-obstacle environment is optimized.

[0005] It is evident that existing technologies suffer from the following problems: the use of a single sensor results in inaccurate reflection of obstacle information in the monitoring data of the surrounding environment; at the same time, the lack of a suitable way to adjust the obstacle avoidance algorithm to cope with different obstacle environments leads to poor obstacle avoidance accuracy and poor safety of humanoid robots. Summary of the Invention

[0006] To address this, the present invention provides a humanoid robot based on a multi-cooperative intelligent algorithm and its control method, which overcomes the problems in the prior art where the use of a single sensor leads to the inability to accurately reflect obstacle information in the monitoring data of the surrounding environment, and the lack of a suitable way to adjust the obstacle avoidance algorithm to cope with different obstacle environments, resulting in poor obstacle avoidance accuracy and poor safety of the humanoid robot.

[0007] To achieve the above objectives, in one aspect, the present invention provides a humanoid robot based on a multi-cooperative intelligent algorithm, comprising:

[0008] Skeletal structure;

[0009] Servo motors are used to drive the joint movements of humanoid robots.

[0010] The sensing module is used to collect real-time video of the humanoid robot's surroundings and the position data of obstacles;

[0011] The data acquisition unit is connected to the sensing module and is used to acquire real-time video of the humanoid robot's surroundings and the location data of obstacles.

[0012] The first analysis unit, which is connected to the data acquisition unit, is used to construct a circle with the humanoid robot as the center and a first preset length as the radius, and to sample the circle at a second preset length as the sampling interval, calculate the risk level evaluation value of each sampling point, and the humanoid robot moves from the center of the circle towards the sampling point with the smallest risk level evaluation value.

[0013] The target detection algorithm is used to detect targets in the latest frame of the real-time video to obtain the number of obstacles, and the adjustment of the first preset length is determined based on the number of obstacles.

[0014] The adjustment of the second preset length is determined based on the distance to the nearest obstacle;

[0015] The second analysis unit is connected to the data acquisition unit and the first analysis unit respectively, and is used to determine the correction of the preset length based on whether the humanoid robot has reached the destination when the humanoid robot stops moving.

[0016] The correction method for adjusting the preset length is determined based on the distance between the humanoid robot and the first adjacent obstacle;

[0017] An adjustment unit, connected to the first analysis unit, is used to determine an adjustment coefficient for adjusting the first preset length based on a first relative difference between the number of obstacles and a preset number, and to determine an adjustment coefficient for adjusting the second preset length based on a second relative difference between the distance between the nearest obstacles and a preset distance between the nearest obstacles.

[0018] A correction unit, which is connected to the second analysis unit, is used to correct the preset length according to the correction coefficient;

[0019] The nearest obstacle distance is the Euclidean distance between the first and second nearest obstacles to the humanoid robot.

[0020] Furthermore, the first analysis unit calculates the risk level assessment value according to the following formula, and sets it as follows:

[0021]

[0022] Where P represents the risk level assessment value, A0 represents the Euclidean distance between the current sampling point and the destination, Ai represents the Euclidean distance between the current sampling point and the i-th obstacle, n represents the number of obstacles, and a is a constant.

[0023] Furthermore, the first analysis unit determines the adjustment of the first preset length based on the comparison result that the number of obstacles in the image is greater than a preset number.

[0024] Furthermore, the adjustment unit determines to adjust the first preset length with a first adjustment coefficient based on the comparison result where the first relative difference is less than or equal to the first preset relative difference;

[0025] Based on the comparison result that the first relative difference is greater than the first preset relative difference, the first preset length is adjusted by the second adjustment coefficient.

[0026] The second adjustment coefficient is less than the first adjustment coefficient.

[0027] Furthermore, the first analysis unit determines the adjustment of the second preset length based on the comparison results of the nearest obstacle spacing being less than or equal to the preset nearest obstacle spacing.

[0028] Furthermore, the adjustment unit determines to adjust the second preset length with a third adjustment coefficient based on the comparison result that the second relative difference is less than or equal to the second preset relative difference;

[0029] Based on the comparison result that the second relative difference is greater than the second preset relative difference, the second preset length is adjusted by the fourth adjustment coefficient.

[0030] The fourth adjustment coefficient is less than the third adjustment coefficient.

[0031] Furthermore, when the humanoid robot stops moving, the second analysis unit determines to correct the preset length based on the determination result that the humanoid robot has not reached its destination.

[0032] Furthermore, the second analysis unit determines, based on the comparison results of the distance being less than or equal to the preset distance, to correct the preset length using a first correction method, wherein the first correction method is to correct the second preset length using a first correction coefficient;

[0033] The correction unit corrects the second preset length according to the following first correction coefficient, setting...

[0034]

[0035] Where Z1 represents the first correction coefficient, L represents the distance between the humanoid robot and the first adjacent obstacle, and L0 represents the preset distance.

[0036] Furthermore, the second analysis unit determines, based on the comparison result that the distance is greater than the preset distance, to correct the preset length using a second correction method, wherein the second correction method is to correct the first preset length using a second correction coefficient;

[0037] The correction unit corrects the first preset length according to the following second correction coefficient, which is set as follows:

[0038]

[0039] Where Z2 represents the second correction coefficient.

[0040] On the other hand, the present invention also provides a control method for a humanoid robot based on a multi-cooperative intelligent algorithm, comprising:

[0041] Construct a circle with the humanoid robot as the center and a first preset length as the radius, and sample the circle at a second preset length as the sampling interval. Calculate the risk level evaluation value of each sampling point, and the humanoid robot moves from the center of the circle towards the sampling point with the lowest risk level evaluation value.

[0042] The target detection algorithm is used to detect targets in the latest frame of the real-time video to obtain the number of obstacles, and the adjustment of the first preset length is determined based on the number of obstacles.

[0043] Alternatively, the adjustment of the second preset length can be determined based on the distance to the nearest obstacle;

[0044] When the humanoid robot stops moving, the preset length is adjusted based on whether the humanoid robot has reached its destination;

[0045] The correction method for adjusting the preset length is determined based on the distance between the humanoid robot and the first adjacent obstacle.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention calculates the risk level evaluation value based on the Euclidean distance between the current sampling point and the destination, the Euclidean distance between the current sampling point and the i-th obstacle, and the number of obstacles to accurately navigate the movement of the humanoid robot. By comprehensively considering the distance of the humanoid robot to the destination and the position of the surrounding obstacles, more reasonable and accurate navigation decisions are made. The humanoid robot can avoid collisions with obstacles while approaching the destination and can flexibly cope with obstacles of different numbers and distributions. The constant 'a' provides an adjustment factor that can adjust the risk level evaluation value according to the distance to the destination to improve the adaptability and accuracy of the risk level evaluation value.

[0047] Furthermore, the present invention determines whether to adjust the first preset length based on the comparison result of the number of obstacles in the image with the preset number, so as to quickly and accurately judge the complexity of the environment, and determines whether it is necessary to adjust the travel length of the humanoid robot so that the humanoid robot takes a more cautious step to avoid collisions, increase safety, and improve the robot's adaptability to complex environments.

[0048] Furthermore, the present invention determines an adjustment coefficient for adjusting the first preset length based on the comparison result of the first relative difference and the first preset relative difference, so as to accurately adjust the step length of the humanoid robot according to the density of obstacles in the environment, which helps the robot maintain a balance between travel efficiency and safety in different environments.

[0049] Furthermore, the present invention determines whether to adjust the second preset length based on the comparison result of the distance between the nearest obstacle and the preset distance between the nearest obstacles, so as to provide a more accurate obstacle avoidance strategy. At the same time, it takes into account the physical size of the humanoid robot, improves the obstacle avoidance accuracy of the humanoid robot when passing through narrow spaces, reduces the risk of collision, improves safety, and enhances the adaptability to complex environments.

[0050] Furthermore, the present invention determines an adjustment coefficient for adjusting the second preset length based on the comparison result of the second relative difference and the second preset relative difference, so as to adopt different obstacle avoidance strategies under different obstacle distribution conditions, thereby improving the accuracy of obstacle avoidance, reducing the risk of collision, and enhancing safety.

[0051] Furthermore, the present invention can determine whether the preset length needs to be modified based on the completion status of the task to improve the flexibility and efficiency during task execution. When the humanoid robot stops moving but has not reached the destination, it indicates that the humanoid robot has encountered a local minimum problem. By modifying the preset length, the problem of the humanoid robot getting stuck in a local minimum and being unable to move can be avoided, thus better coping with complex or changing environments and enhancing robustness.

[0052] Furthermore, the present invention determines a correction method for adjusting the preset length based on the comparison result between the distance and the preset distance, thereby accurately improving the adaptability and obstacle avoidance ability of the humanoid robot in complex environments, while avoiding getting stuck in local minima and being unable to move.

[0053] Furthermore, the present invention calculates a first correction coefficient according to a formula to accurately correct the second preset length, avoiding insufficient correction requiring multiple corrections that would reduce travel efficiency, and avoiding excessive correction that would waste computing resources.

[0054] Furthermore, the present invention calculates a second correction coefficient according to a formula to accurately correct the first preset length, avoiding insufficient correction that requires multiple corrections and thus reduces travel efficiency, and avoiding over-correction that could lead to collisions with the humanoid robot. Attached Figure Description

[0055] Figure 1 This is a first axonometric drawing of a humanoid robot based on a multi-cooperative intelligent algorithm according to an embodiment of the present invention.

[0056] Figure 2 This is a second axonometric drawing of a humanoid robot based on a multi-cooperative intelligent algorithm according to an embodiment of the present invention.

[0057] Figure 3 This is a schematic diagram of the structure of a humanoid robot based on a multi-cooperative intelligent algorithm according to an embodiment of the present invention;

[0058] Figure 4 This is a flowchart illustrating the control method of a humanoid robot based on a multi-cooperative intelligent algorithm, as described in an embodiment of the present invention.

[0059] In the diagram, 1-skeleton structure, 2-servo motor, 3-first ultra-wide-angle camera, 4-humanoid robot head, 5-second ultra-wide-angle camera, 6-first millimeter-wave radar, and 7-second millimeter-wave radar. Detailed Implementation

[0060] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0061] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0062] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0063] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0064] Please see Figure 1-4 As shown, Figure 1 This is a first axonometric drawing of a humanoid robot based on a multi-cooperative intelligent algorithm according to an embodiment of the present invention. Figure 2 This is a second axonometric drawing of a humanoid robot based on a multi-cooperative intelligent algorithm according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of a humanoid robot based on a multi-cooperative intelligent algorithm according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating the control method of a humanoid robot based on a multi-cooperative intelligent algorithm, as described in an embodiment of the present invention.

[0065] This invention relates to a humanoid robot based on a multi-cooperative intelligent algorithm, comprising:

[0066] Skeletal structure 1;

[0067] Servo motor 2 is used to drive the joint movements of the humanoid robot;

[0068] The sensing module includes a first ultra-wide-angle camera 3 positioned in front of the humanoid robot's head 4 and a second ultra-wide-angle camera 5 positioned behind the humanoid robot's head 4, for collecting real-time video of the humanoid robot's surroundings.

[0069] The sensing module also includes a first millimeter-wave radar 6 located to the right of the first ultra-wide-angle camera 3 and a second millimeter-wave radar 7 located to the right of the second ultra-wide-angle camera 5, for collecting position data of obstacles around the humanoid robot.

[0070] The data acquisition unit is connected to the sensing module to acquire real-time video of the humanoid robot's surroundings and the location data of obstacles.

[0071] The first analysis unit, which is connected to the data acquisition unit, is used to construct a circle with the humanoid robot as the center and a first preset length C1 as the radius, and to sample the circle with a second preset length C2 as the sampling interval, calculate the risk level evaluation value P of each sampling point, and the humanoid robot moves from the center of the circle to the sampling point with the smallest risk level evaluation value.

[0072] The target detection algorithm is used to detect targets in the latest frame of the real-time video to obtain the number of obstacles W, and the adjustment of the first preset length C1 is determined based on the number of obstacles W.

[0073] The adjustment of the second preset length C2 is determined based on the distance J between the nearest obstacles;

[0074] The second analysis unit is connected to the data acquisition unit and the first analysis unit respectively, and is used to determine the correction of the preset length based on whether the humanoid robot has reached the destination when the humanoid robot stops moving.

[0075] An adjustment unit, connected to the first analysis unit, is used to determine an adjustment coefficient for adjusting the first preset length C1 based on a first relative difference ΔW between the number of obstacles W and a preset number W0, and to determine an adjustment coefficient for adjusting the second preset length C2 based on a second relative difference ΔJ between the distance between the nearest obstacles J and a preset distance between the nearest obstacles J0.

[0076] A correction unit, which is connected to the second analysis unit, is used to correct the preset length according to the correction coefficient;

[0077] Wherein, the nearest obstacle distance J is the Euclidean distance between the first and second nearest obstacles of the humanoid robot.

[0078] In this embodiment of the invention, the target detection algorithm is preferably YOLOv5, the value of the first preset length C1 is preferably 150% of the width of the humanoid robot, the value of the second preset length C2 is preferably 5% of the circumference of the circle, and the width is the shoulder width of the humanoid robot. Those skilled in the art can adjust the first preset length C1 and the second preset length C2 according to specific circumstances.

[0079] Specifically, the first analysis unit calculates the risk level assessment value P according to the following formula, and sets:

[0080]

[0081] Where A0 represents the Euclidean distance between the current sampling point and the destination, Ai represents the Euclidean distance between the current sampling point and the i-th obstacle, n represents the number of obstacles, and a is a constant.

[0082] Specifically, this invention calculates a risk level assessment value based on the Euclidean distance between the current sampling point and the destination, the Euclidean distance between the current sampling point and the i-th obstacle, and the number of obstacles to accurately navigate the movement of the humanoid robot. By comprehensively considering the distance from the humanoid robot to the destination and the position of surrounding obstacles, it makes more reasonable and accurate navigation decisions. The humanoid robot can avoid collisions with obstacles while approaching the destination and can flexibly cope with obstacles of different numbers and distributions. The constant 'a' provides an adjustment factor that can adjust the risk level assessment value according to the distance from the destination to improve the adaptability and accuracy of the risk level assessment value.

[0083] Specifically, the first analysis unit determines whether to adjust the first preset length C1 based on the comparison result between the number W of obstacles in the image and the preset number W0;

[0084] If W≤W0, then the first analysis unit determines not to adjust the first preset length C1;

[0085] If W > W0, then the first analysis unit determines to adjust the first preset length C1;

[0086] In this embodiment of the invention, the preset quantity W0 is 3. Those skilled in the art can adjust the preset quantity W0 according to specific circumstances.

[0087] Specifically, the present invention determines whether to adjust the first preset length based on the comparison between the number of obstacles in the image and the preset number, so as to quickly and accurately judge the complexity of the environment, and determines whether it is necessary to adjust the travel length of the humanoid robot so that the humanoid robot takes a more cautious step to avoid collisions, increase safety, and improve the robot's adaptability to complex environments.

[0088] Specifically, the adjustment unit calculates the first relative difference ΔW between the number of obstacles W and the preset number W0 under the condition determined by the first analysis unit for adjusting the first preset length C1, and determines the adjustment coefficient for adjusting the first preset length C1 based on the comparison result of the first relative difference ΔW and the first preset relative difference ΔW0, and sets ΔW = (W-W0) / W0.

[0089] If △W≤△W0, then the adjustment unit determines to adjust the first preset length C1 with the first adjustment coefficient X1;

[0090] If △W>△W0, then the adjustment unit determines to adjust the first preset length C1 with the second adjustment coefficient X2;

[0091] Set the adjusted first preset length to Cc1 = C1 × X1 or Cc1 = C1 × X2.

[0092] Where 0 < X2 < X1 < 1, in this embodiment of the invention, the value of the first adjustment coefficient X1 is preferably 0.95, and the value of the second adjustment coefficient X2 is preferably 0.85.

[0093] In this embodiment of the invention, the first preset relative difference △W0 is 1. The first preset relative difference △W0 is obtained when the number of obstacles W is 6. Those skilled in the art can adjust the first adjustment coefficient X1 and the second adjustment coefficient X2 according to the specific circumstances.

[0094] Specifically, the present invention determines an adjustment coefficient for adjusting the first preset length based on the comparison result between the first relative difference and the first preset relative difference, so as to accurately adjust the step length of the humanoid robot according to the density of obstacles in the environment, which helps the robot maintain a balance between travel efficiency and safety in different environments.

[0095] Specifically, when the adjustment unit determines to adjust the first preset length C1 with the second adjustment coefficient X2, the first analysis unit determines whether to adjust the second preset length C2 based on the comparison result between the nearest obstacle distance J and the preset nearest obstacle distance J0, wherein the nearest obstacle distance J is the Euclidean distance between the first and second nearest obstacles of the humanoid robot.

[0096] If J≤J0, then the first analysis unit determines to adjust the second preset length C2;

[0097] If J > J0, then the first analysis unit determines not to adjust the second preset length C2;

[0098] In this embodiment of the invention, the preset distance J0 between the nearest obstacles is 200% of the width of the humanoid robot. Those skilled in the art can adjust the preset distance J0 between the nearest obstacles according to specific circumstances.

[0099] Specifically, the present invention determines whether to adjust the second preset length based on the comparison between the distance to the nearest obstacle and the preset distance to the nearest obstacle, so as to provide a more accurate obstacle avoidance strategy. At the same time, it takes into account the physical size of the humanoid robot, improves the obstacle avoidance accuracy of the humanoid robot when passing through narrow spaces, reduces the risk of collision, improves safety, and enhances the adaptability to complex environments.

[0100] Specifically, under the condition that the first analysis unit determines that the second preset length C2 is to be adjusted, the adjustment unit calculates the second relative difference ΔJ between the nearest obstacle distance J and the preset nearest obstacle distance J0, and determines the adjustment coefficient for adjusting the second preset length C2 based on the comparison result between the second relative difference ΔJ and the second preset relative difference ΔJ0, and sets ΔJ = (J0 - J) / J0.

[0101] If △J≤△J0, then the adjustment unit determines to adjust the second preset length C2 with the third adjustment coefficient X3;

[0102] If △J>△J0, then the adjustment unit determines to adjust the second preset length C2 with the fourth adjustment coefficient X4;

[0103] Set the adjusted first preset length to Cc2 = C2 × X3 or Cc2 = C2 × X4.

[0104] Where 0 < X4 < X3 < 1, in this embodiment of the invention, the value of the third adjustment coefficient X3 is preferably 0.7, and the value of the fourth adjustment coefficient X4 is preferably 0.5.

[0105] In this embodiment of the invention, the second preset relative difference △J0 is 0.25. The second preset relative difference △J0 is obtained when the distance J between the nearest obstacles is 150% of the width of the humanoid robot. Those skilled in the art can adjust the first adjustment coefficient X1 and the second adjustment coefficient X2 according to the specific circumstances.

[0106] Specifically, the present invention determines an adjustment coefficient for adjusting the second preset length based on the comparison result of the second relative difference and the second preset relative difference, so as to adopt different obstacle avoidance strategies under different obstacle distribution conditions, thereby improving the accuracy of obstacle avoidance, reducing the risk of collision, and enhancing safety.

[0107] Specifically, when the humanoid robot stops moving, the second analysis unit determines whether to correct the preset length based on whether the humanoid robot has reached its destination.

[0108] If the humanoid robot reaches its destination, the second analysis unit determines that the preset length should not be corrected.

[0109] If the humanoid robot fails to reach its destination, the second analysis unit determines to correct the preset length.

[0110] Specifically, the present invention can determine whether the preset length needs to be modified based on the completion status of the task in order to improve the flexibility and efficiency of the task execution process. When the humanoid robot stops moving but has not reached the destination, it indicates that the humanoid robot has encountered a local minimum problem. By modifying the preset length, the problem of the humanoid robot getting stuck in a local minimum and being unable to move can be avoided, thus better coping with complex or changing environments and enhancing robustness.

[0111] Specifically, the second analysis unit determines the distance L between the humanoid robot and the first adjacent obstacle under the condition of correcting the preset length, and determines the correction method for correcting the preset length based on the comparison result of the distance L and the preset distance L0.

[0112] If L≤L0, then the second analysis unit determines to correct the preset length using the first correction method;

[0113] If L > L0, then the second analysis unit determines to correct the preset length using the second correction method;

[0114] The first correction method is to correct the second preset length C2 with a first correction coefficient Z1, and the second correction method is to correct the first preset length C1 with a second correction coefficient Z2.

[0115] In this embodiment of the invention, the preset distance L0 is set to 300% of the width of the humanoid robot. Those skilled in the art can adjust the preset distance L0 according to specific circumstances.

[0116] Specifically, the present invention determines a correction method for adjusting the preset length based on the comparison result between the distance and the preset distance, thereby accurately improving the adaptability and obstacle avoidance ability of the humanoid robot in complex environments, while avoiding getting stuck in local minimum points and being unable to move.

[0117] Specifically, under the condition that the second analysis unit determines to correct the preset length in the first correction method, the correction unit corrects the second preset length C2 according to the following first correction coefficient Z1, and sets:

[0118]

[0119] The corrected second preset length is set to Cd2 = C2 × Z1.

[0120] Specifically, the present invention calculates a first correction coefficient according to a formula to accurately correct the second preset length, avoiding insufficient correction that requires multiple corrections and thus reduces travel efficiency, and avoiding excessive correction that leads to a waste of computing resources.

[0121] Specifically, under the condition that the second analysis unit determines to correct the preset length in a second correction manner, the correction unit corrects the first preset length C1 according to the following second correction coefficient Z2, which is set as follows:

[0122]

[0123] The corrected second preset length is set to Cd1 = C1 × Z2.

[0124] Specifically, the present invention calculates a second correction coefficient according to a formula to accurately correct the first preset length, avoiding insufficient correction that requires multiple corrections and thus reduces travel efficiency, and avoiding over-correction that could lead to collisions with the humanoid robot.

[0125] This invention relates to a control method for a humanoid robot based on a multi-cooperative intelligent algorithm, comprising:

[0126] Step S1: Construct a circle with the humanoid robot as the center and a first preset length C1 as the radius, and sample the circle with a second preset length C2 as the sampling interval. Calculate the risk level evaluation value P for each sampling point. The humanoid robot moves from the center of the circle towards the sampling point with the smallest risk level evaluation value.

[0127] Step S2: Perform target detection on the latest frame of the real-time video using the target detection algorithm to obtain the number of obstacles W, and determine the adjustment of the first preset length C1 based on the number of obstacles W.

[0128] Alternatively, the adjustment of the second preset length C2 can be determined based on the distance J between the nearest obstacles;

[0129] Step S3: When the humanoid robot stops moving, determine the correction of the preset length based on whether the humanoid robot has reached the destination;

[0130] Step S4: Determine the correction method for the preset length based on the distance L between the humanoid robot and the first adjacent obstacle.

[0131] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0132] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A humanoid robot based on multi-collaborative intelligence algorithm, characterized in that, The application relates to a robot control system, which comprises: a skeleton structure; a servo motor for driving the joint movement of a humanoid robot; a sensing module for collecting real-time video and position data of obstacles around the humanoid robot; a data acquisition unit connected with the sensing module for acquiring real-time video and position data of obstacles around the humanoid robot; a first analysis unit connected with the data acquisition unit for constructing a circle with the humanoid robot as the center and a first preset length as the radius, sampling the circle with a second preset length as the sampling interval, calculating the risk degree evaluation value of each sampling point, and moving the humanoid robot in the direction of the sampling point with the minimum risk degree evaluation value; detecting the number of obstacles in the latest image of the real-time video according to a target detection algorithm, and adjusting the first preset length according to the number of obstacles; adjusting the second preset length according to the distance between the nearest obstacles; a second analysis unit connected with the data acquisition unit and the first analysis unit for determining the modification of the preset length according to whether the humanoid robot reaches the destination when the humanoid robot stops moving; determining the modification mode of the preset length according to the distance between the humanoid robot and the first nearest obstacle; an adjustment unit connected with the first analysis unit for determining the adjustment coefficient of the first preset length according to the first relative difference between the number of obstacles and a preset number, and determining the adjustment coefficient of the second preset length according to the second relative difference between the distance between the nearest obstacles and a preset distance between the nearest obstacles; a modification unit connected with the second analysis unit for modifying the preset length according to the modification coefficient. The distance between the nearest obstacles is the Euclidean distance between the first nearest obstacle and the second nearest obstacle of the humanoid robot.

2. The multi-collaborative intelligence algorithm based humanoid robot according to claim 1, wherein, The first analysis unit calculates the risk degree evaluation value according to the following formula, and sets: Wherein, P represents the risk degree evaluation value, A0 represents the Euclidean distance between the current sampling point and the destination, Ai represents the Euclidean distance between the current sampling point and the ith obstacle, n represents the number of obstacles, and a is a constant, 3. The multi-cooperative intelligence algorithm based humanoid robot according to claim 2, wherein, The first analysis unit determines the adjustment of the first preset length based on the comparison result that the number of obstacles in the image is greater than a preset number.

4. The multi-cooperative intelligence algorithm based humanoid robot according to claim 3, wherein, The adjustment unit determines the adjustment of the first preset length with the first adjustment coefficient based on the comparison result that the first relative difference is less than or equal to a first preset relative difference. The adjustment unit determines the adjustment of the first preset length with the second adjustment coefficient based on the comparison result that the first relative difference is greater than the first preset relative difference. The second adjustment coefficient is less than the first adjustment coefficient.

5. The multi-collaborative intelligence algorithm based humanoid robot according to claim 4, wherein, The first analysis unit determines the adjustment of the second preset length based on the comparison result that the distance between the nearest obstacles is less than or equal to a preset distance between the nearest obstacles.

6. The multi-cooperative intelligence algorithm based humanoid robot according to claim 5, wherein, The adjustment unit determines the adjustment of the second preset length with the third adjustment coefficient based on the comparison result that the second relative difference is less than or equal to a second preset relative difference. The adjustment unit determines the adjustment of the second preset length with the fourth adjustment coefficient based on the comparison result that the second relative difference is greater than the second preset relative difference. The fourth adjustment coefficient is less than the third adjustment coefficient.

7. The multi-cooperative intelligent algorithm based humanoid robot according to claim 6, wherein, The second analysis unit determines to correct the preset length based on a result of determining that the humanoid robot does not reach the destination under the condition that the humanoid robot stops moving.

8. The multi-cooperative intelligent algorithm based humanoid robot according to claim 7, wherein, The second analysis unit determines to correct the preset length in a first correction mode based on a result of comparing the distance with the preset distance, the first correction mode being to correct the second preset length by a first correction coefficient. The correction unit corrects the second preset length according to the first correction coefficient, which is set as wherein Z1 represents the first correction coefficient, L represents the distance between the humanoid robot and the first adjacent obstacle, and L0 represents the preset distance.

9. The multi-cooperative intelligent algorithm based humanoid robot according to claim 8, wherein, The second analysis unit determines to correct the preset length in a second correction mode based on a result of comparing the distance with the preset distance, the second correction mode being to correct the first preset length by a second correction coefficient. The correction unit corrects the first preset length according to the second correction coefficient, which is set as wherein Z2 represents the second correction coefficient.

10. A control method of a humanoid robot based on a multi-collaborative intelligent algorithm, according to any one of claims 1-9, characterized in that, a circle is constructed with the humanoid robot as the center and the first preset length as the radius, and the circle is sampled with the second preset length as the sampling interval, the risk degree evaluation value of each sampling point is calculated, and the humanoid robot moves in the direction of the center pointing to the sampling point with the minimum risk degree evaluation value; a target detection algorithm is used to detect the number of obstacles in the latest frame of the real-time video, and the adjustment of the first preset length is determined according to the number of obstacles; or, the adjustment of the second preset length is determined according to the distance between the closest obstacles; when the humanoid robot stops moving, the correction of the preset length is determined according to whether the humanoid robot reaches the destination; the correction mode of the preset length is determined according to the distance between the humanoid robot and the first adjacent obstacle.

Images