Bionic robot action control method based on behavior control

Abstract

The invention discloses a biomimetic robot action control method based on behavior control, and relates to the technical field of biomimetic robot control, and the method comprises the following steps: building a global terrain coordinate system, completing multi-sensor calibration, enabling a robot to start an autonomous navigation task, and enabling a visual sensor to start to collect a surrounding environment image; processing the acquired image, and extracting color and texture features in the image; an interferent recognition module recognizes interferents in the data and extracts feature information of the interferents; the interference index analysis module calculates an interference index according to the feature information of the interferent; the terrain environment complexity evaluation module collects terrain environment information in real time and evaluates the complexity of the terrain environment; the action control decision module synthesizes the interference index and the terrain environment complexity evaluation result, makes an action control decision and sends a control instruction to the motion execution mechanism, and the problems of motion intelligence and safety of the bionic robot in the complex environment are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of bionic robot control, and specifically provides a motion control method for a bionic robot based on behavior control. Background Art

[0002] With the continuous development of robot technology, bionic robots are gradually widely used in fields such as disaster rescue, complex terrain inspection, military reconnaissance, and field operations. Compared with traditional wheeled or tracked mobile robots, bionic robots simulate the limb structures and movement patterns of animals, and have stronger terrain adaptability and flexibility, especially suitable for task execution in complex environments such as rugged, uneven, and obstacle-dense terrains. However, in complex natural environments, robots face multiple challenges such as obstacle avoidance, path selection, terrain perception and adaptation. It is difficult to achieve intelligent response to the environment simply relying on rule presets or path planning algorithms.

[0003] Currently, some bionic robots use a method combining path planning and obstacle avoidance control for navigation control, but often ignore the comprehensive consideration of multi-dimensional factors such as the type, influence degree of interference objects in the environment, and the complexity of the terrain environment itself. For example, when the robot faces interference objects such as grass, stones, and fallen leaves, the system often can only make a simple judgment based on the presence or absence of obstacles, lacking in-depth analysis of the specific characteristics of obstacles and selection of response strategies. At the same time, most current bionic robot control strategies are based on rigid control logic, lacking the behavior decision-making ability similar to that of living organisms, and it is difficult to dynamically adjust action strategies according to environmental changes, resulting in low motion efficiency and safety in uncertain environments. For example, when encountering vision occlusion or potential pits, traditional robots lack a tentative action and perception linkage mechanism, and there are risks such as misstepping and falling. Therefore, there is an urgent need for a bionic robot control method with behavior decision-making ability, which can identify the characteristics of interference objects in the environment, evaluate their interference degree, and dynamically adjust action control strategies in combination with terrain complexity analysis. Summary of the Invention

[0004] The purpose of the present invention is to provide a motion control method for a bionic robot based on behavior control to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: A motion control method for a bionic robot based on behavior control, including the following steps:

[0006] Step S1: Establish a global terrain coordinate system, complete multi-sensor calibration, the robot starts an autonomous navigation task, and the vision sensor starts to collect images of the surrounding environment;

[0007] Step S2: Process the collected images and extract color and texture features in the images;

[0008] Step S3: The interference recognition module identifies the interference therein and extracts its feature information;

[0009] Step S4: The interference index analysis module calculates the interference index according to the feature information of the interference;

[0010] Step S5: The terrain environment complexity assessment module collects terrain environment information in real time and assesses the complexity of the terrain environment;

[0011] Step S6: The motion control decision-making module comprehensively considers the interference index and the terrain environment complexity assessment result, makes a motion control decision, and sends a control instruction to the motion execution mechanism;

[0012] Step S7: The motion execution mechanism drives the limb joint motors of the robot according to the control instruction to achieve corresponding actions;

[0013] Step S8: During the movement of the robot, the above steps are continuously repeated to adjust the motion control strategy in real time to ensure that the robot can safely and accurately complete the autonomous navigation task.

[0014] According to the above technical solution, in the step S2, the specific method for processing the collected image and extracting the color and texture features in the image includes:

[0015] Step S21: Convert the color image collected by the camera into a grayscale image, and perform grayscale processing using the weighted average method. The calculation formula is: Gray = 0.299R + 0.587G + 0.114B, where R, G, and B are the pixel values of the red, green, and blue channels in the color image respectively, and Gray is the pixel value of the grayscale image;

[0016] Step S22: Use the Gaussian filtering algorithm to denoise the grayscale image. Let the Gaussian kernel be G(x, y), and its calculation formula is where σ is the standard deviation of the Gaussian kernel, (x, y) is the coordinate of the pixel. Convolve the Gaussian kernel with the image to obtain the denoised image;

[0017] Step S23: Convert the grayscale image into the HSV color space, and its conversion formula is:

[0018] V = max(R, G, B);

[0019]

[0020] Step S24: Perform histogram statistics on the hue component in the HSV color space, divide the value range of the hue into several regions, count the number of pixels in each region to obtain the color histogram, and extract the main color features in the image according to the color histogram;

[0021] Step S25: For each pixel in the image, take it as the center and select a 3×3 neighborhood. Use the gray value of the central pixel of the neighborhood as the threshold, and compare it with the gray values of the surrounding 8 neighborhood pixels. When the gray value of the neighborhood pixel is greater than or equal to the gray value of the central pixel, mark the position of this neighborhood pixel as 1; otherwise, mark it as 0. Repeat this step until each pixel obtains an 8-bit binary number, and convert it to a decimal number as the LBP value of this pixel;

[0022] Step S26: Statistically analyze the distribution of different LBP values in the image to obtain the LBP histogram; and extract the image texture features according to the LBP histogram.

[0023] According to the above technical solution, in step S3, the specific method for the interference object recognition module to recognize the interference objects therein and extract their feature information is as follows:

[0024] Step S31: Obtain the image after image preprocessing, color feature extraction, and texture feature extraction;

[0025] Step S32: Establish a template database of interference objects, and sort out and summarize samples of different angles and illumination conditions of each template in the template database;

[0026] Step S33: Adopt a template matching algorithm to divide the image into multiple sub-regions, compare each sub-region in the recognition image with the templates in the template library, calculate the similarity between the sub-region and the template, and the similarity is measured by the normalized cross-correlation coefficient. The formula is where I(x,y) is the pixel value of the sub-region, is the average pixel value of the sub-region, T(x,y) is the pixel value of the template, is the average pixel value of the template. When the similarity exceeds the threshold G u then it is determined that there is an interference object in this sub-region, and the interference object is preliminarily classified according to the similarity matching result;

[0027] Step S34: For the recognized interference object region, extract its feature information; the feature information includes the color feature, texture feature of the interference object, and the position and size features of the interference object in the image.

[0028] According to the above technical solution, in step S4, the specific method for the interference index analysis module to calculate the interference index according to the feature information of the interference object is as follows:

[0029] Step S41: Assign corresponding weights to each feature information according to the influence degree of the interference object on the robot's autonomous navigation task;

[0030] Step S42: Through the calculation formula of the interference index The calculated interference index Q, where Q represents the interference index, ω i represents the weight of the i-th feature, f i represents the value of the i-th feature, and n represents the total number of features;

[0031] Step S43: According to the calculated interference index Q, evaluate the influence degree of the interference object; set an interference index threshold Q0, and classify the interference objects into different levels. When Q ≤ Q0, it is judged as low interference, and when Q > Q0, it is judged as high interference.

[0032] According to the above technical solution, in step S5, the terrain environment complexity evaluation module collects terrain environment information in real time. The specific workflow for evaluating the complexity of the terrain environment includes the following steps:

[0033] Step S51: Obtain the image information processed through the previous steps, including the color histogram, LBP histogram, and the interference object recognition result;

[0034] Step S52: Extract the color histogram entropy H c and the complexity index W f of the texture corresponding to the LBP histogram;

[0035] Step S53: Calculate the terrain environment complexity index F d = αH c + βW f , where α and β are the control parameters of the color histogram entropy and the complexity index of the texture corresponding to the LBP histogram respectively; d Step S54: When the terrain environment complexity index F

[0036] > F d >, it is judged that the current terrain environment is complex. When the terrain environment complexity index F d0 ≤ F d ≤ F d0 , it is judged that the current terrain environment is simple.

[0037] According to the above technical solution, the specific action control decision in step S6 is as follows:

[0038] When the interference index is at a high interference level and the terrain environment is complex, the robot chooses to bypass;

[0039] When the interference index is at a low interference level and the terrain environment is simple, the robot directly crosses or tramples on the interference object;

[0040] When the interference index is at a low interference level but the terrain environment is complex, further analyze and determine whether to consider that the interference object blocks the potential pit hazard. If it is determined that there is a potential pit hazard, handle it with caution; otherwise, pass normally.

[0041] According to the above technical solution, the specific method for analyzing and determining whether to consider that the interference object blocks the potential pit hazard is as follows:

[0042] Step A: Based on the depth map data of the vision sensor, use the gradient threshold method to extract the terrain elevation mutation area. If the depth difference between adjacent pixels exceeds ΔD, mark it as a suspected pit boundary;

[0043] Step B: Combine the infrared thermal imaging data to analyze the surface temperature anomaly area. If the temperature gradient distribution differs from the surrounding terrain by more than ΔT, it is determined as a potential pit;

[0044] Step C: Project the detected pit features onto the global terrain coordinate system, and record the pit center coordinate set {C k (x k ,y k ,z k )}, where k = 1, 2,..., n;

[0045] Step D: Calculate the three-dimensional space distance between adjacent pits Traverse all pit combinations to generate a distance matrix;

[0046] Step E: Calculate the pit distribution density index based on the distance matrix When ρ > ρ0, it is determined that it is necessary to consider that the interference object blocks the potential pit hazard, where ρ0 is the preset pit distribution density threshold.

[0047] According to the above technical solution, the method of handling with caution if it is determined that there is a potential pit hazard is as follows:

[0048] Control the robot foot end to make a tentative contact at the edge of the interference object;

[0049] Real-time monitor the ground reaction force fluctuation through the six-axis force sensor. If the vertical stiffness coefficient K v < K v0 then it is determined that there is a risk of virtual support;

[0050] Combine the piezoelectric film sensor to detect the surface vibration frequency spectrum. When the main frequency component offset Δf > Δf max it is confirmed that the pit exists.

[0051] Compared with the prior art, the beneficial effects achieved by the present invention are: the present invention

[0052] (1) By introducing a joint evaluation mechanism of the interference index and the terrain complexity index, the robot can dynamically adjust its action strategy according to environmental changes, realizing diverse behavioral options such as crossing, detouring, or tentative passage, significantly reducing the risk of misjudging obstacles or falling;

[0053] (2) Through image processing technologies such as Gaussian filtering, HSV color space analysis, and LBP texture coding, more discriminative visual features are extracted, effectively improving the recognition accuracy of small obstacles or camouflaged interference objects;

[0054] (3) Through the integration of multi-dimensional perception information such as image color and texture feature extraction, interference object recognition and feature analysis, and terrain environment complexity assessment, the bionic robot is given a complete control loop of "perception-analysis-decision-execution", significantly improving its adaptability in unknown and complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0056] In the attached figure:

[0057] Figure 1 This is a flow chart of a bionic robot motion control method based on behavior control of the present invention. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0059] See also Figure 1 The present invention provides a technical solution: a bionic robot motion control method based on behavior control, comprising the following steps:

[0060] Step S1: Establish a global terrain coordinate system, complete multi-sensor calibration, the robot starts the autonomous navigation task, and the visual sensor begins to collect images of the surrounding environment;

[0061] Step S2: Process the collected image to extract color and texture features in the image;

[0062] The specific methods for processing the collected images and extracting the color and texture features in the images include:

[0063] Step S21: Convert the color image captured by the camera into a grayscale image. The weighted average method is used for grayscale processing. The calculation formula is: Gray = 0.299R + 0.587G + 0.114B, where R, G, and B are the pixel values of the red, green, and blue channels in the color image respectively, and Gray is the pixel value of the grayscale image. This can reduce the amount of data in the image while retaining the main brightness information of the image, facilitating subsequent processing.

[0064] Step S22: Use the Gaussian filtering algorithm to denoise the grayscale image. Let the Gaussian kernel be G(x, y), and its calculation formula is where σ is the standard deviation of the Gaussian kernel, (x, y) are the coordinates of the pixel. Convolve the Gaussian kernel with the image to obtain the denoised image.

[0065] Step S23: Convert the grayscale image into the HSV color space. The HSV color space consists of three components: hue, saturation, and brightness. Compared with the RGB color space, it is more in line with human perception of colors. The conversion formula is:

[0066] V = max(R, G, B);

[0067]

[0068] Step S24: Perform a histogram statistics on the hue component in the HSV color space. Divide the value range of the hue into several regions, count the number of pixels in each region to obtain the color histogram, and extract the main color features in the image according to the color histogram.

[0069] Step S25: For each pixel in the image, take it as the center and select a 3×3 neighborhood. Use the grayscale value of the neighborhood center pixel as the threshold and compare it with the grayscale values of the surrounding 8 neighborhood pixels. When the grayscale value of the neighborhood pixel is greater than or equal to the grayscale value of the center pixel, mark the position of the neighborhood pixel as 1, otherwise mark it as 0. Repeat this step until each pixel obtains an 8-bit binary number, and convert it into a decimal number as the LBP value of the pixel.

[0070] Step S26: Statistically analyze the distribution of different LBP values in the image to obtain the LBP histogram; and extract the image texture features according to the LBP histogram.

[0071] By introducing grayscale processing, Gaussian filtering for noise reduction, HSV color space conversion, and a texture feature extraction method based on local binary pattern (LBP), a complete, stable, and efficient image preprocessing and feature extraction process is constructed. This method can not only accurately extract the color and texture information of images in complex environments, but also has good anti-noise ability and computational efficiency, significantly improving the environmental perception accuracy and robustness of bionic robots. Compared with existing image processing technologies, the HSV color space adopted in the present invention is closer to human visual perception, and the LBP texture descriptor has stronger local feature expression ability, enabling the system to maintain stable operation when facing challenges such as light changes and terrain complexity and providing a solid data foundation for subsequent behavior decision-making, with significant practical value and promotion prospects.

[0072] Step S3: The interference object recognition module recognizes the interference objects therein and extracts their feature information;

[0073] The specific method for the interference object recognition module to recognize the interference objects therein and extract their feature information is as follows:

[0074] Step S31: Obtain the image after image preprocessing, color feature extraction, and texture feature extraction;

[0075] Step S32: Establish a template database of interference objects, and sort out and summarize samples of different angles and lighting conditions of each template in the template database;

[0076] Step S33: Adopt a template matching algorithm, divide the image into multiple sub-regions, compare each sub-region in the recognition image with the templates in the template library, calculate the similarity between the sub-region and the template, and the similarity is measured by the normalized cross-correlation coefficient. The formula is where I(x, y) is the pixel value of the sub-region, is the average pixel value of the sub-region, T(x, y) is the pixel value of the template, is the average pixel value of the template. When the similarity exceeds the threshold G u , it is determined that there is an interference object in the sub-region, and the interference object is preliminarily classified according to the similarity matching result, such as categories like big stones, plastic bags, dead branches and fallen leaves, etc.;

[0077] Step S34: For the recognized interference object region, extract its feature information; the feature information includes the color feature of the interference object (obtained from the color histogram), the texture feature (obtained from the LBP histogram), as well as the position and size features of the interference object in the image;

[0078] By constructing a database of interferer templates and combining the preprocessing results of color and texture features, a template matching algorithm is introduced for regional similarity analysis, thereby effectively identifying the types and distribution characteristics of interferers in images. The normalized cross-correlation coefficient (NCC) is used as the similarity measurement criterion, which can accurately identify interferers in complex environments such as different lighting and angles, and preliminarily classify them into categories such as large stones, plastic bags, or withered branches and fallen leaves. On this basis, further extract the feature information such as the color, texture, position, and size of the interferers, providing accurate and comprehensive environmental cognition basis for the obstacle avoidance behavior and path planning of the bionic robot in complex environments. Compared with the traditional method based on single feature or rough detection, the method of the present invention has higher recognition accuracy and environmental adaptability, significantly enhancing the processing ability and intelligent level of the bionic robot for interference factors in dynamic natural environments.

[0079] Step S4: The interference index analysis module calculates the interference index according to the feature information of the interferer;

[0080] The specific method for the interference index analysis module to calculate the interference index according to the feature information of the interferer is as follows:

[0081] Step S41: Assign corresponding weights to each feature information according to the influence degree of the interferer on the robot's autonomous navigation task;

[0082] Step S42: Through the calculation formula of the interference index The calculated interference index Q, where Q represents the interference index, ω i represents the weight of the i-th feature, f i represents the value of the i-th feature, and n represents the total number of features;

[0083] Step S43: Evaluate the influence degree of the interferer according to the calculated interference index Q; set an interference index threshold Q0, and classify the interferers into different levels. When Q ≤ Q0, it is judged as low interference, and when Q > Q0, it is judged as high interference.

[0084] Step S5: The terrain environment complexity evaluation module collects terrain environment information in real time and evaluates the complexity of the terrain environment;

[0085] The specific workflow for the terrain environment complexity evaluation module to collect terrain environment information in real time and evaluate the complexity of the terrain environment includes the following steps:

[0086] Step S51: Obtain the image information processed through the previous steps, including the color histogram, LBP histogram, and interferer recognition results;

[0087] Step S52: Extract the color histogram entropy H according to the preset algorithm c, the complexity index W of the texture corresponding to the LBP histogram f ;

[0088] Step S53: Through the terrain environment complexity index formula F d = αH c + βW f , calculate the terrain environment complexity index F d , where α and β are the control parameters of the color histogram entropy and the complexity index of the texture corresponding to the LBP histogram respectively;

[0089] Step S54: When the terrain environment complexity index F d > F d0 , it is determined that the current is a complex terrain environment. When the terrain environment complexity index F d ≤F d0 , it is determined that the current is a simple terrain environment; In terms of terrain environment complexity assessment, by fusing the color histogram and texture feature (LBP histogram) information of the image, introducing two quantitative indicators of color entropy and texture complexity, and constructing a weighted calculation model, a terrain environment complexity index is formed. This index comprehensively reflects the multi-dimensional information complexity of the current terrain environment by setting the weight parameters α and β, and realizes the intelligent discrimination of whether the environment is complex or not by comparing with the threshold. This method can evaluate the changes and complexity of the terrain where the robot is located in real time, provide an important basis for the adjustment of subsequent motion control strategies, and significantly improve the adaptability and autonomous behavior decision-making ability of the robot in a dynamic complex environment.

[0090] Step S6: The motion control decision-making module comprehensively considers the interference index and the terrain environment complexity assessment result, makes a motion control decision, and sends the control instruction to the motion execution mechanism;

[0091] The specific content of the motion control decision is:

[0092] When the interference index is at a high interference level and the terrain environment is complex, the robot chooses to bypass;

[0093] When the interference index is at a low interference level and the terrain environment is simple, the robot directly crosses or tramples on the interference object;

[0094] When the interference index is at a low interference level but the terrain environment is complex, further analyze and judge whether it is considered that the interference object blocks the potential pit danger. If it is judged that there is a potential pit danger, handle it carefully, otherwise pass normally.

[0095] The specific method for analyzing and judging whether it is considered that the interference object blocks the potential pit danger is:

[0096] Step A: Based on the depth map data of the vision sensor, the gradient threshold method is used to extract the terrain elevation mutation area. If the depth difference between adjacent pixels exceeds ΔD, it is marked as a suspected pit boundary;

[0097] Step B: Combining the infrared thermal imaging data, analyze the surface temperature anomaly area. If the temperature gradient distribution differs from the surrounding terrain by more than ΔT, it is determined as a potential pit;

[0098] Step C: Project the detected pit features onto the global terrain coordinate system, and record the set of pit center coordinates {C k (x k ,y k ,z k )}, where k = 1, 2, …, n;

[0099] Step D: Calculate the three-dimensional spatial distance between adjacent pits Traverse all pit combinations to generate a distance matrix;

[0100] Step E: Calculate the pit distribution density index based on the distance matrix When ρ > ρ0, it is judged that it is necessary to consider that the interference object blocks the potential pit danger, where ρ0 is the preset pit distribution density threshold.

[0101] If it is judged that there is a potential pit danger, the cautious handling method is:

[0102] Control the robot foot end to make a tentative contact at the edge of the interference object;

[0103] Real-time monitor the ground reaction force fluctuation through the six-axis force sensor. If the vertical stiffness coefficient K v < K v0 Then it is determined that there is a risk of virtual support;

[0104] Combine the piezoelectric film sensor to detect the surface vibration spectrum. When the main frequency component offset Δf > Δf max Confirm the existence of the pit; In terms of motion control decision-making, an intelligent strategy system based on the joint evaluation of the interference index and terrain complexity is constructed. For this reason, the system extracts the terrain mutation area through the visual depth map, combines the infrared thermal imaging to detect the surface temperature anomaly, identifies the pit position and conducts spatial analysis, calculates the pit distribution density to determine the occlusion risk. Once it is judged that there is a potential pit, the robot will execute a cautious handling strategy, including the foot end making a tentative contact with the ground, the six-axis force sensor detecting the support stability and the piezoelectric film sensor monitoring the surface vibration spectrum, to confirm the existence of virtual support or pit, effectively improving the stability and safety of the robot in uncertain terrain.

[0105] Step S7: The motion execution mechanism drives the limb joint motors of the robot according to the control instructions to implement corresponding actions, such as walking, striding, detouring, etc.;

[0106] Step S8: During the movement of the robot, the above steps are continuously repeated to adjust the motion control strategy in real time to ensure that the robot can safely and accurately complete the autonomous navigation task.

[0107] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0108] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0110] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the present invention and the claims. All of these fall within the protection scope of the present invention.

Claims

1. A bionic robot motion control method based on behavior control, characterized in that: It includes the following steps: Step S1: Establish a global terrain coordinate system, complete multi-sensor calibration, the robot starts the autonomous navigation task, and the vision sensor begins to collect images of the surrounding environment; Step S2: Process the collected images and extract color and texture features in the images; Step S3: The interference object recognition module recognizes the interference objects therein and extracts their feature information; Step S4: The interference index analysis module calculates the interference index according to the feature information of the interference objects; Step S5: The terrain environment complexity evaluation module collects terrain environment information in real time and evaluates the complexity of the terrain environment; Step S6: The motion control decision-making module comprehensively considers the interference index and the evaluation result of the terrain environment complexity, makes a motion control decision, and sends the control instruction to the motion execution mechanism; Step S7: The motion execution mechanism drives the limb joint motors of the robot according to the control instruction to achieve corresponding actions; Step S8: During the movement of the robot, continuously repeat the above steps, and adjust the motion control strategy in real time to ensure that the robot can safely and accurately complete the autonomous navigation task.

2. The method for controlling the actions of a bionic robot based on behavior control according to claim 1, wherein: In the above-mentioned step S2, the specific method for processing the collected images and extracting color and texture features in the images includes: Step S21: Convert the color image collected by the camera into a grayscale image, and perform grayscale processing using the weighted average method. The calculation formula is: Gray = 0.299R + 0.587G + 0.114B, where R, G, and B are the pixel values of the red, green, and blue channels in the color image respectively, and Gray is the pixel value of the grayscale image; Step S22: Denoise the grayscale image using the Gaussian filtering algorithm. Let the Gaussian kernel be G(x, y), and its calculation expression is where σ is the standard deviation of the Gaussian kernel, (x, y) are the coordinates of the pixel, and convolve the Gaussian kernel with the image to obtain the denoised image; Step S23: Convert the grayscale image into the HSV color space, and its conversion formula is: V = max(R, G, B); Step S24: Perform histogram statistics on the hue component in the HSV color space, divide the value range of the hue into several regions, count the number of pixels in each region, obtain the color histogram, and extract the main color features in the image according to the color histogram; Step S25: For each pixel in the image, take it as the center and select a 3×3 neighborhood. Use the grayscale value of the neighborhood center pixel as the threshold and compare it with the grayscale values of the surrounding 8 neighborhood pixels. When the grayscale value of the neighborhood pixel is greater than or equal to the grayscale value of the center pixel, mark the position of the neighborhood pixel as 1, otherwise mark it as 0. Repeat this step until each pixel obtains an 8-bit binary number, and convert it into a decimal number as the LBP value of the pixel; Step S26: Statistically analyze the distribution of different LBP values in the image to obtain the LBP histogram; and extract the image texture features according to the LBP histogram.

3. The method for controlling the actions of a bionic robot based on behavior control according to claim 1, wherein: In the above-mentioned step S3, the specific method for the interference object recognition module to recognize the interference objects therein and extract their feature information is: Step S31: Obtain the image after image preprocessing, color feature extraction, and texture feature extraction; Step S32: Establish a template database of interference objects, and sort out and summarize samples of different angles and lighting conditions of each template in the template database; Step S33: Using the template matching algorithm, divide the image into multiple sub-regions, compare each sub-region in the recognition image with the templates in the template library, calculate the similarity between the sub-region and the template, and the similarity is measured by the normalized cross-correlation coefficient. The formula is where I(x,y) is the pixel value of the sub-region, is the average pixel value of the sub-region, T(x,y) is the pixel value of the template, is the average pixel value of the template. When the similarity exceeds the threshold G u it is determined that there are interfering objects in the sub-region, and the interfering objects are preliminarily classified according to the similarity matching results; Step S34: Extract the feature information of the recognized interference object area; The feature information includes the color feature, texture feature of the interference object, as well as the position and size features of the interference object in the image.

4. A bionic robot motion control method based on behavior control according to claim 1, characterized in that: In the step S4, the specific method for the interference index analysis module to calculate the interference index according to the feature information of the interference object is as follows: Step S41: Assign corresponding weights to each feature information according to the influence degree of the interference object on the robot's autonomous navigation task; Step S42: Through the calculation formula of the interference index The interference index Q is calculated, where Q represents the interference index, ω i represents the weight of the i-th feature, f i represents the value of the i-th feature, and n represents the total number of features; Step S43: Evaluate the influence degree of the interference object according to the calculated interference index Q; Set an interference index threshold Q0, and classify the interference objects into different levels. When Q ≤ Q0, it is judged as low interference, and when Q > Q0, it is judged as high interference.

5. A bionic robot motion control method based on behavior control according to claim 1, characterized in that: In the step S5, the terrain environment complexity evaluation module collects terrain environment information in real time, and the specific work process for evaluating the complexity of the terrain environment includes the following steps: Step S51: Obtain the image information processed by the previous steps, including the color histogram, LBP histogram, and the interference object recognition result; Step S52: Extract the color histogram entropy H according to a preset algorithm c and the complexity index W of the texture corresponding to the LBP histogram f ; Step S53: Through the terrain environment complexity index formula F d = αH c + βW f , calculate the terrain environment complexity index F d , where α and β are the control parameters of the complexity indexes of the color histogram entropy and the texture corresponding to the LBP histogram respectively; Step S54: When the terrain environment complexity index F d >F d0 it is determined that the current situation is a complex terrain environment. When the terrain environment complexity index F d ≤F d0 it is determined that the current situation is a simple terrain environment.

6. A method for controlling the actions of a bionic robot based on behavior control according to claim 1, characterized in that: The specific content of the action control decision in the step S6 is as follows: When the interference index is at a high interference level and the terrain environment is complex, the robot chooses to bypass; When the interference index is at a low interference level and the terrain environment is simple, the robot directly crosses or tramples on the interference object; When the interference index is at a low interference level but the terrain environment is complex, further analyze and judge whether it is considered that the interference object blocks the potential pit danger. If it is judged that there is a potential pit danger, handle it carefully, otherwise pass normally.

7. A method for controlling the actions of a bionic robot based on behavior control according to claim 6, characterized in that: The specific method for analyzing and judging whether it is considered that the interference object blocks the potential pit danger is as follows: Step A: Based on the depth map data of the vision sensor, use the gradient threshold method to extract the terrain elevation mutation area. If the depth difference between adjacent pixels exceeds ΔD, mark it as the suspected pit boundary; Step B: Combine the infrared thermal imaging data to analyze the surface temperature abnormal area. If the temperature gradient distribution is different from the surrounding terrain by more than ΔT, it is determined as a potential pit; Step C: Project the detected pit features onto the global terrain coordinate system, and record the set of pit center coordinates {C k (x k ,y k ,z k )}, where k = 1, 2, …, n; Step D: Calculate the three-dimensional spatial distance between adjacent pits Traverse all pit combinations to generate a distance matrix; Step E: Calculate the pit distribution density index based on the distance matrix When ρ > ρ0, it is determined that it is necessary to consider that the interference object blocks the potential pit danger, where ρ0 is a preset pit distribution density threshold.

8. The method for controlling the actions of a bionic robot based on behavior control according to claim 6, characterized in that: The way to handle it carefully if it is judged that there is a potential pit danger is as follows: Control the robot's foot end to make a tentative contact at the edge of the interference object; Real-time monitoring of the ground reaction force fluctuations through a six-axis force sensor. If the vertical stiffness coefficient K v < K v0 then it is determined that there is a risk of virtual support; Combined with a piezoelectric thin film sensor to detect the surface vibration spectrum, when the offset of the main frequency component Δf > Δf max it is confirmed that the pit exists.

Images