A humanoid robot fusion system with autonomous navigation

Abstract

The present application relates to the field of humanoid robots, and especially relates to a humanoid robot fusion system with autonomous navigation, which comprises a man-machine interaction module, a data acquisition module, a data fusion processing module, a navigation module and a motion control module, the man-machine interaction module is used for receiving user instructions, the data acquisition module is used for collecting environmental data near the humanoid robot, the data fusion processing module is used for fusing and processing the collected data, the navigation module is used for obtaining risk indexes of various obstacles in the environment and generating a moving path of the humanoid robot according to the risk indexes and the user instructions, and the motion control module is used for controlling the humanoid robot, the present application sets the risk indexes, is favorable for analyzing the risk conditions of various obstacles photographed, removes the obstacles with smaller risks from the obstacle avoidance process, and thus reduces the complexity of obstacle avoidance calculation.

Description

Technical Field

[0001] This invention relates to the field of humanoid robots, and more particularly to a humanoid robot fusion system with autonomous navigation. Background Technology

[0002] In recent years, with the rapid development of artificial intelligence, robotics, sensing technology, and automatic control, humanoid robots, as an important branch of service robots, have gradually become a research hotspot. Due to their anthropomorphic structure, humanoid robots are more easily adapted to human living and working environments, possessing good environmental adaptability and human-computer interaction potential. Especially in scenarios such as home service, medical care, disaster relief, and industrial collaboration, humanoid robots have shown broad application prospects.

[0003] For example, the prior art disclosed in CN118906050A discloses a humanoid robot navigation method, apparatus, computer device, and storage medium. The method includes: acquiring a semantic map; acquiring traversable and impassable areas based on the map tile information of each map tile; acquiring the humanoid robot's task, navigation start point, and navigation end point; dividing the task into multiple sub-tasks; planning multiple reference work processes based on the traversable and impassable areas, navigation start point, navigation end point, and multiple sub-tasks; scoring each reference work process; obtaining the optimal work process based on the score; planning the path between adjacent optimal work points based on the traversable and impassable areas and the real-time map, as the optimal movement path; and enabling the humanoid robot to navigate according to the optimal work process and optimal movement path.

[0004] Another typical prior art method for humanoid robot navigation based on a hybrid fuzzy embedded PID control algorithm, such as CN119596675A, includes: measuring distance; using infrared sensors to measure the distance between the humanoid robot and the forward, right, and left obstacles, defining fuzzy control variables and determining their value ranges; defining the fuzzy control variables as the distance to the forward obstacle, the distance to the right obstacle, the distance to the left obstacle, and the optimized turning angle, and determining their value ranges, and generating fuzzy membership functions.

[0005] Let's look at an existing technology, such as CN114474054A, which discloses a humanoid robot navigation method. Based on a robot system, it includes a robot and several movable ranging devices. The robot is equipped with a robotic arm and several fixed sensors. The navigation method includes a device-grabbing process and a path planning process. The device-grabbing process involves the robotic arm receiving a grasping command, grasping the specified type of ranging device, adaptively adjusting the device's position, and matching the corresponding SLAM algorithm based on the type of ranging device. The path planning process involves the robot combining map information, information about surrounding obstacles obtained from sensors and / or ranging devices, the robot's real-time position, and the target point's position to generate a path to the target point based on a SLAM algorithm. During the robot's movement, its position is monitored in real-time to determine if it deviates from the path, and the path is adjusted if it does.

[0006] Currently, existing humanoid robot navigation systems have a rather one-sided view of obstacles and usually adopt a complete avoidance approach during navigation, which makes obstacle avoidance calculations quite complex. In order to solve the common problems in this field, this invention was made. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of current systems by proposing a humanoid robot fusion system with autonomous navigation.

[0008] To overcome the shortcomings of the prior art, the present invention adopts the following technical solution:

[0009] A humanoid robot fusion system with autonomous navigation includes a human-computer interaction module, a data acquisition module, a data fusion processing module, a navigation module, and a motion control module. The human-computer interaction module receives user commands. The data acquisition module collects environmental data near the humanoid robot according to user commands. The data fusion processing module fuses and processes the data collected by the data acquisition module. The navigation module obtains risk indicators of various obstacles in the environment and generates a movement path for the humanoid robot based on the risk indicators and user commands. The motion control module controls the humanoid robot according to the movement path generated by the navigation module.

[0010] Furthermore, the human-computer interaction module includes an input unit, an instruction generation unit, and a display unit. The input unit is used to receive user input, the instruction generation unit is used to generate corresponding user instructions based on the user input, and the display unit is used to display various working parameters of the humanoid robot.

[0011] Furthermore, the data acquisition module includes multiple cameras for capturing environmental images near the humanoid robot; the data fusion processing module includes a preprocessing unit, a mapping unit, and a map data storage unit. The preprocessing unit is used to reduce noise in the captured environmental images, the mapping unit is used to generate a current environmental map based on the preprocessed images, and the map data storage unit is used to store the images generated by the mapping unit.

[0012] Furthermore, the navigation module includes a judgment unit, an algorithm storage unit, and a path planning unit. The algorithm storage unit is used to store various algorithms required for path planning. The path planning unit is used to derive the algorithms from the algorithm storage unit and generate the humanoid robot's movement path based on user instructions and the environmental map. The judgment unit is used to obtain the risk indicators of each obstacle in the environmental map and determine whether each obstacle needs to be considered in the path planning. The algorithm storage unit is also used to store obstacle avoidance algorithms, which are used to adjust the movement path in real time according to the obstacle situation.

[0013] Furthermore, the motion control module includes an instruction analysis unit and a gait generation unit. The instruction analysis unit is used to analyze user instructions and the final movement path of the navigation module and convert the analysis results into the form of input to the gait generation unit. The gait generation unit is used to generate control signals based on the output of the instruction analysis unit and control the gait of the humanoid robot through the control signals.

[0014] Furthermore, the system's workflow includes the following steps:

[0015] S1, the human-computer interaction module receives user commands;

[0016] S2, the data acquisition module collects environmental data around the humanoid robot;

[0017] S3, the data fusion and processing module fuses and processes the data collected by the data acquisition module to obtain an environmental map;

[0018] S4, the navigation module generates the humanoid robot's movement path;

[0019] S5, the motion control module controls the humanoid robot according to the generated movement path and user instructions.

[0020] Furthermore, generating the humanoid robot's movement path includes the following steps:

[0021] S41, the path planning unit generates the initial path of the humanoid robot based on user instructions and the environment map;

[0022] S42, During the movement of the humanoid robot, the judgment unit obtains the risk indicators of each obstacle in the map based on the pre-processed image;

[0023] S43, the path planning unit avoids obstacles with a risk index greater than 1 according to the obstacle avoidance algorithm, and obtains the adjusted movement path;

[0024] S44, input the adjusted movement path into the motion control module.

[0025] The beneficial effects achieved by this invention are: 1. By setting risk indicators, it is beneficial to analyze the risk status of each obstacle captured, remove obstacles with lower risk from the obstacle avoidance process, thereby reducing the complexity of obstacle avoidance calculation and avoiding situations where there is no way to go due to too many small obstacles with lower risk.

[0026] 2. Replacing obstacles with spherical obstacles makes it easier to analyze various situations that may occur when a humanoid robot steps on an obstacle, thereby obtaining the maximum tilt angle that the humanoid robot may experience when encountering the obstacle during movement, which is beneficial for unified calculation of various obstacles. Attached Figure Description

[0027] The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on illustrating the principles of the embodiments. In different views, the same reference numerals designate the same parts.

[0028] Figure 1 This is a schematic diagram of the structure of the present invention.

[0029] Figure 2 This is a flowchart of the process of the present invention.

[0030] Figure 3 This is a flowchart illustrating the process of generating the movement path of the humanoid robot according to the present invention.

[0031] Figure 4 This diagram illustrates the relationship between the maximum tilt angle that the humanoid robot of the present invention may encounter when it moves and the maximum cross-sectional area of ​​the spherical obstacle corresponding to the obstacle, and the ground contact area of ​​the humanoid robot's feet. Detailed Implementation

[0032] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated beforehand. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0033] Example 1: According to Figure 1 , Figure 2 , Figure 3 and Figure 4 This embodiment provides a humanoid robot fusion system with autonomous navigation, including a human-computer interaction module, a data acquisition module, a data fusion processing module, a navigation module, and a motion control module. The human-computer interaction module is used to receive user commands. The data acquisition module is used to collect environmental data near the humanoid robot according to user commands. The data fusion processing module is used to fuse and process the data collected by the data acquisition module. The navigation module is used to obtain risk indicators of various obstacles in the environment based on the data processed by the data fusion processing module and generate a movement path for the humanoid robot based on the risk indicators and user commands. The motion control module is used to control the humanoid robot according to the movement path generated by the navigation module.

[0034] Furthermore, the human-computer interaction module includes an input unit, an instruction generation unit, and a display unit. The input unit is used to receive user input, the instruction generation unit is used to generate corresponding user instructions based on the user input, and the display unit is used to display various working parameters of the humanoid robot.

[0035] Specifically, the user instructions include, but are not limited to, the humanoid robot's destination and movement speed.

[0036] Furthermore, the data acquisition module includes multiple cameras for capturing environmental images near the humanoid robot; the data fusion processing module includes a preprocessing unit, a mapping unit, and a map data storage unit. The preprocessing unit is used to reduce noise in the captured environmental images, the mapping unit is used to generate a current environmental map based on the denoised images, and the map data storage unit is used to store the map generated by the mapping unit.

[0037] Specifically, the mapping unit generates an environment map using the VSLAM algorithm.

[0038] Furthermore, the navigation module includes a judgment unit, an algorithm storage unit, and a path planning unit. The algorithm storage unit is used to store various algorithms required for path planning. The path planning unit is used to derive the algorithms from the algorithm storage unit and generate the humanoid robot's movement path based on user instructions and the environmental map. The judgment unit is used to obtain the risk indicators of each obstacle in the environmental map and determine whether each obstacle needs to be considered in the path planning. The algorithm storage unit is also used to store obstacle avoidance algorithms, which are used to adjust the movement path in real time according to the obstacle situation.

[0039] Furthermore, the motion control module includes an instruction analysis unit and a gait generation unit. The instruction analysis unit is used to analyze the final movement path of the navigation module and convert the analysis results into the form of input to the gait generation unit. The gait generation unit is used to generate control signals based on the output of the instruction analysis unit and control the gait of the humanoid robot through the control signals.

[0040] Furthermore, the system's workflow includes the following steps:

[0041] S1, the human-computer interaction module receives user commands;

[0042] S2, the data acquisition module collects environmental data around the humanoid robot;

[0043] S3, the data fusion and processing module fuses and processes the data collected by the data acquisition module to obtain an environmental map;

[0044] S4, the navigation module generates the humanoid robot's movement path;

[0045] S5, the motion control module controls the humanoid robot according to the generated movement path.

[0046] Furthermore, generating the humanoid robot's movement path includes the following steps:

[0047] S41, the path planning unit generates the initial path of the humanoid robot based on user instructions and the environment map;

[0048] S42, During the movement of the humanoid robot, the judgment unit obtains the risk indicators of each obstacle in the map based on the pre-processed image;

[0049] Specifically, before calculating risk indicators, it is necessary to perform image recognition on obstacles. Through the steps of point cloud generation, center point calculation, and sphere modeling, the image of the obstacle in the environment map is replaced with a spherical obstacle (after replacement, there may be a phenomenon where part of the spherical obstacle is located below the ground surface, which does not affect the actual calculation results). The center of the spherical obstacle is the center point of the obstacle, and its radius is equal to the distance from the point of the obstacle farthest from the center point to the center point.

[0050] Specifically, the risk index of a certain obstacle can be calculated using the following formula:

[0051] ;

[0052] ;

[0053] ;

[0054] in, This is a risk indicator used to characterize the risk of a humanoid robot falling over when it steps on an obstacle. The higher the indicator, the greater the risk. The volume of the obstacle as identified by image recognition technology. The volume threshold is set by a person skilled in the art based on the maximum hemispherical obstacle volume that a humanoid robot can accept (one that can barely pass through without falling). Let L be the diameter of the spherical obstacle, L be the length of the shortest side of the quadrilateral footplate of the circular robot, and e be the natural constant. This represents the maximum cross-sectional area of ​​the spherical obstacle's portion on the ground surface. This represents the maximum cross-sectional area of ​​the obstacle in the captured image, that is, the maximum value of its cross-sectional area obtained from all images of the obstacle. The maximum tilt angle that the feet of a humanoid robot can withstand. This refers to the maximum tilt angle generated by the foot of a humanoid robot when it steps on the corresponding spherical obstacle during movement. This is the volume influence function.

[0055] in, , satisfy , The area ratio affects the weight. The tilt angle affects the weight. The weighting of volume influence is determined by the experience of those skilled in the art, and one possible method for setting the weighting is as follows: =0.1, =0.5 and =0.4, its setting is based on: The corresponding item is This is used to measure the risk to robot movement caused by changes in the area of ​​an obstacle after processing, by measuring its cross-sectional area. Existing modeling techniques are relatively mature, therefore... Set to minimum; The corresponding item is The proportion of this tilt angle directly reflects the risk of the robot tilting due to the height of the obstacle, and should be given the highest weight. Set it to 0.5. The corresponding item is It is used to measure whether the volume of an obstacle is close to a threshold, indirectly reflecting the robot's tilt risk, and its importance is second only to... Therefore, it is set to 0.4. The above weight settings are only examples; those skilled in the art can flexibly design according to the above value rules in specific applications.

[0056] The ratio setting for the maximum cross-sectional area aims to determine image sharpness and is set exponentially. This is because existing modeling and image recognition technologies are already quite mature. and The difference is generally close to 0, that is Generally less than At this point, the value of the exponent will not be too large. If the difference is too large, it proves that there may be a large error in the modeling part or the image recognition part, which will greatly affect the subsequent judgment. Therefore, it is necessary to increase the impact of the difference in the form of an exponent. The proportion directly reflects the risk of the robot tilting due to the height of the obstacle, and setting it in proportion is more intuitive. The proportion indirectly reflects the robot's tilt risk, when and When less than 1, and Their contributions to risk indicators are roughly the same; both are linear functions at this point. When they are greater than 1 (for...),... That is, the maximum possible tilt angle that can be accepted. That is, the identified volume is greater than the volume threshold, which needs to be reflected. Compare The angular proportion contributes more to the risk indicators, therefore the angular proportion is set as a linear function while the volume proportion is set as a logarithmic function. The logarithm is chosen with base e to ensure... The contribution to logarithmic growth will be neither too fast nor too slow; this is achieved by internally setting the logarithmic function to... This can guarantee the piecewise function It is a continuous function.

[0057] Specifically, the volume threshold is set by those skilled in the art based on the foot contact area of ​​the humanoid robot and its actual performance; the maximum tilt angle that the humanoid robot can accept is obtained by those skilled in the art based on performance experiments on the humanoid robot; by setting risk indicators, it is beneficial to analyze the risk of each obstacle captured, remove obstacles with lower risk from the obstacle avoidance process, thereby reducing the complexity of obstacle avoidance calculations and avoiding situations where there is no way to go due to too many small obstacles with lower risk; generally, by considering This allows for the assessment of obstacle risk. However, during actual movement, the tilt angle varies depending on the position of the robot's foot on the obstacle, and it's impossible for it to always be the maximum tilt angle. Considering only the maximum tilt angle to obtain the risk indicator can easily lead to an inflated risk index. Therefore, the identified volume V and a volume threshold are incorporated into the risk indicator calculation, and these are compared with... Weighting the data helps to indirectly assess obstacle risk through volume, reducing the possibility of inflated risk indicators. Furthermore, since the system heavily relies on image recognition and modeling technologies, introducing weighted data further enhances its effectiveness. This helps to take into account the risks caused by modeling errors and image recognition errors in the actual risks.

[0058] like Figure 4 As shown, Figure 4 This is a diagram showing the relationship between the maximum tilt angle that a humanoid robot may encounter when it moves, the maximum cross-sectional area of ​​the corresponding spherical obstacle, and the ground contact area of ​​the humanoid robot's feet.

[0059] S43, the path planning unit avoids obstacles with risk indicators greater than or equal to the risk indicator threshold according to the obstacle avoidance algorithm, and obtains the adjusted movement path.

[0060] Specifically, the risk indicator threshold can be set based on the risk indicator under critical conditions. When the risk indicator exceeds the risk indicator threshold, it is considered that the robot will fall if it steps on the corresponding obstacle. Greater than At that time, it was believed that there were risks in the image recognition or modeling process. = For the critical case, when Greater than At that time, it was believed that the robot posed a risk of falling. For the critical case, when Greater than At that time, it was believed that the robot posed a risk of falling. This is a critical situation, where the risk indicator threshold value is:

[0061] ;

[0062] S44, input the adjusted movement path into the motion control module.

[0063] The beneficial effects of this solution are: 1. By setting risk indicators, it is beneficial to analyze the risk of each obstacle captured in the video, remove obstacles with lower risk from the obstacle avoidance process, thereby reducing the complexity of obstacle avoidance calculation and avoiding situations where there is no way to go due to too many small obstacles with lower risk.

[0064] 2. Replacing obstacles with spherical obstacles makes it easier to analyze various situations that may occur when a humanoid robot steps on an obstacle, thereby obtaining the maximum tilt angle that the humanoid robot may experience when encountering the obstacle during movement, which is beneficial for unified calculation of various obstacles.

[0065] Example 2: This example should be understood as a further improvement on the aforementioned example, and it also includes a preferred replacement setting method for the risk index threshold. In S43, the path planning unit avoids obstacles with risk indices greater than or equal to the risk index threshold according to the obstacle avoidance algorithm. This solution proposes a preferred replacement setting method for the risk index threshold, thereby screening obstacles according to the actual environmental conditions and reducing the problem of high robot wear rate caused by the threshold setting in Example 1.

[0066] The optimized risk indicator threshold can be calculated using the following formula:

[0067] ;

[0068] in, To optimize risk indicator thresholds, The risk index threshold is set according to the method proposed in Example 1. When the risk index is greater than the risk index threshold YZ, it is considered that the robot will fall when it steps on the corresponding obstacle. When the risk index ZB is between [YYZ, YZ], it is considered that the robot will cause unnecessary wear when it steps on the corresponding obstacle. e is a natural constant, and A is the volume V identified in the environmental map that is less than the volume threshold. The number of obstacles, Let a be the recognition volume of the a-th obstacle. Let A be the average recognition volume of the obstacles. The total area occupied by all obstacles. The total area occupied by the humanoid robot.

[0069] This embodiment, based on Embodiment 1, reduces the risk index threshold YZ according to the environment, thus obtaining an optimized risk index threshold YZZ. In Embodiment 1, obstacle avoidance was only required when the risk index was greater than the risk index threshold YZ, because when the risk index was greater than the risk index threshold YZ, the robot was considered to have a high probability of falling. In this embodiment, by setting YZZ, the risk index threshold is reduced to obtain an optimized risk index threshold, allowing the robot to avoid obstacles even when the probability of falling is very small, thereby reducing the wear on the robot's feet when stepping on obstacles. An exponential function is set, and according to the formula, the value of the exponential function ranges from 0 to 1. Multiplying YZ by this exponential function can achieve the effect of reducing the threshold; YZZ is set according to the total area occupied by the humanoid robot's activity range. Enlarging and shrinking the obstacle allows for more obstacle avoidance in larger spaces. For obstacles with a high risk index but below the risk index threshold YZ, although not avoiding them will not cause the robot to fall, stepping on them will still cause some wear and tear. By considering the footprint to reduce the risk index threshold YZ and thus obtaining the optimal risk index threshold YZZ, it is beneficial to reduce the impact of such obstacles on the robot.

[0070] Furthermore, considering that wear and tear is the reason for reducing the risk indicator threshold, and ground area is a limiting factor in reducing the risk indicator threshold, in order to reduce wear and tear, the threshold needs to be reduced as much as possible to enable more obstacle avoidance. However, if the threshold is too low, it is not feasible in areas with limited traversable space, which may result in the entire area being filled with obstacles to avoid, leaving no way to go or requiring a long detour to reach the destination. Therefore, the area occupied is considered to reduce the occurrence of situations where there is no way to go or a long detour. When the optimized risk indicator threshold YZZ is small, it may be due to the large area occupied by obstacles, or it may be due to the standard deviation. There are many large obstacles, and although the smaller YZZ requires more obstacle avoidance, thus increasing the robot's path, it results in less wear on the robot's feet compared to directly stepping on the obstacles without obstacle avoidance. It's worth noting that since YZZ is smaller than YZ, the method used in Example 1 for determining the obstacles to be avoided is the same in Example 2.

[0071] Furthermore, by considering the standard deviation of the obstacle volume... When the standard deviation is large, it is assumed that the obstacle volume is less uniform and larger obstacles are more likely to appear. By introducing the standard deviation into the exponential function to reduce the risk index threshold YZ, an optimized risk index threshold YZZ is obtained. This can not only avoid larger obstacles but also avoid some smaller obstacles, and increase the number of obstacle avoidance attempts, thereby reducing the wear and tear on the robot when passing through obstacles.

[0072] The beneficial effects of this embodiment are: by considering the standard deviation of the footprint and volume to optimize the risk index threshold, an optimized risk index threshold is obtained, which is beneficial to reduce the impact of obstacles on the robot in larger spaces.

[0073] The above-disclosed content is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the description and drawings of the present invention are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops. The above units are merely examples, and those skilled in the art can adopt corresponding units according to actual needs when implementing this solution.

Claims

1. A humanoid robot fusion system with autonomous navigation, characterized by, The system comprises a human-computer interaction module, a data acquisition module, a data fusion processing module, a navigation module and a motion control module, the human-computer interaction module is used for receiving user instructions, the data acquisition module is used for collecting environmental data near the humanoid robot according to the user instructions, the data fusion processing module is used for fusing and processing the collected data of the data acquisition module, the navigation module is used for obtaining risk indicators of various obstacles in the environment and generating a moving path of the humanoid robot according to the risk indicators and the user instructions, and the motion control module is used for controlling the humanoid robot according to the moving path generated by the navigation module. The risk indicator of a certain obstacle is calculated according to the following formula: ； ； ； wherein, is a risk index, used to represent the risk of the humanoid robot falling down when stepping on the obstacle, the greater the index, the greater the risk, is the volume of the obstacle identified by image recognition technology, is a volume threshold, which is set by a person skilled in the art according to the maximum hemispherical obstacle volume that can be accepted by the humanoid robot, is the diameter of the spherical obstacle, L is the length of the shortest side of the quadrilateral foot plate of the circular robot, and e is a natural constant, is the maximum cross-sectional area of the part of the spherical obstacle corresponding to the obstacle on the ground surface, is the maximum cross-sectional area of the obstacle in the captured image, i.e., the maximum value of the cross-sectional area obtained in all images of the obstacle, is the maximum tilting angle that can be accepted by the foot plate of the humanoid robot, is the maximum tilting angle of the foot plate when the humanoid robot steps on the spherical obstacle corresponding to the obstacle during movement, is a volume influence function; , satisfies , is an area proportion influence weight, is a tilting angle influence weight, is a volume influence weight, and the above weight values are set by a person skilled in the art according to experience.

2. The humanoid robot fusion system with autonomous navigation according to claim 1, wherein, The human-computer interaction module comprises an input unit, an instruction generation unit and a display unit, the input unit is used for receiving user input, the instruction generation unit is used for generating corresponding user instructions according to the user input, and the display unit is used for displaying various working parameters of the humanoid robot.

3. The humanoid robot fusion system with autonomous navigation according to claim 1, wherein, The data acquisition module comprises a plurality of cameras, the cameras are used for shooting environmental images near the humanoid robot, the data fusion processing module comprises a preprocessing unit, a mapping unit and a map data storage unit, the preprocessing unit is used for denoising the shot environmental images, the mapping unit is used for generating a current environmental map according to the preprocessed images, and the map data storage unit is used for storing the images generated by the mapping unit.

4. The humanoid robot fusion system with autonomous navigation according to claim 1, wherein, The navigation module comprises a judgment unit, an algorithm storage unit and a path planning unit, the algorithm storage unit is used for storing various algorithms required for path planning, the path planning unit is used for exporting the algorithms in the algorithm storage unit and generating a moving path of the humanoid robot according to the user instructions and the environmental map, the judgment unit is used for obtaining the risk indicators of various obstacles in the environmental map and judging whether the various obstacles need to be considered in the path planning, and the algorithm storage unit is also used for storing an obstacle avoidance algorithm, the obstacle avoidance algorithm is used for adjusting the moving path in real time according to the obstacle situation.

5. The humanoid robot fusion system with autonomous navigation according to claim 1, wherein, The motion control module comprises an instruction analysis unit and a gait generation unit, the instruction analysis unit is used for analyzing the user instructions and the final moving path of the navigation module and converting the analysis result into the form of input of the gait generation unit, and the gait generation unit is used for generating a control signal according to the output of the instruction analysis unit and controlling the gait of the humanoid robot through the control signal.

6. The humanoid robot fusion system with autonomous navigation of claim 1, wherein, The working process of the system comprises the following steps: S1, the human-computer interaction module receives user instructions; S2, the data acquisition module collects environmental data near the humanoid robot; S3, the data fusion processing module fuses and processes the collected data of the data acquisition module to obtain an environmental map; S4, the navigation module generates a moving path of the humanoid robot; S5, the motion control module controls the humanoid robot according to the generated moving path and the user instructions.

7. The humanoid robot fusion system with autonomous navigation of claim 1, wherein, Generating the moving path of the humanoid robot comprises the following steps: S41, the path planning unit generates an initial path of the humanoid robot according to the user instructions and the environmental map; S42, in the moving process of the humanoid robot, the judgment unit obtains the risk indexes of each obstacle in the map according to the preprocessed image; S43, the path planning unit performs obstacle avoidance on the obstacle with the risk index greater than 1 according to an obstacle avoidance algorithm, to obtain an adjusted moving path; S44, the adjusted moving path is input into the motion control module.

Images