Low-cost combined navigation method and system for quadruped robot in complex environment

Abstract

The invention provides a low-cost combined navigation method and system for a quadruped robot in a complex environment, and belongs to the technical field of robots, and the method specifically comprises the following steps: determining the speed information of the quadruped robot by adopting a speed fitting function according to the walking stride and frequency of the quadruped robot, three attitude angles are output in real time based on an inertial measurement unit installed on the quadruped robot, course error and magnetic declination are compensated through the attitude angles to obtain motion attitude information of the quadruped robot, and navigation position calculation is performed by using information of the quadruped robot at the previous moment, motion attitude information of the quadruped robot at the current moment and speed information. And calculating the position information at the current moment, performing attitude control on the quadruped robot based on the position information and the compensation data in the current preset time period, and updating the navigation position in real time until the positioning information of which the quality meets the requirement is obtained, so as to correct the position error accumulated during navigation position calculation. And the reliability of operation control of the robot is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of robots, and particularly relates to a low-cost combined navigation method and system for a quadruped robot in a complex environment. Background Art

[0002] In order to achieve the navigation control of a quadruped robot, in the existing technical solutions, the combined navigation of lidar / IMU / GNSS or the combined navigation of depth camera / IMU / GNSS can provide high-quality navigation information, but there are problems of relatively high sensor hardware cost and large computational complexity. It is difficult for a quadruped robot to directly perform autonomous navigation based on reinforcement learning motion control without perception, and a high-precision navigation system needs to be installed, which directly increases the system cost and complexity.

[0003] In view of the above technical problems, the present invention provides a low-cost combined navigation method and system for a quadruped robot in a complex environment. Summary of the Invention

[0004] To achieve the object of the present invention, the present invention adopts the following technical solutions:

[0005] According to one aspect of the present invention, a low-cost combined navigation method for a quadruped robot in a complex environment is provided.

[0006] A low-cost combined navigation method for a quadruped robot in a complex environment, characterized in that:

[0007] S1: Real-time obtain the positioning information of the quadruped robot, and when the signal quality of the positioning information does not meet the requirements, proceed to the next step;

[0008] S2: Take the time period when the signal quality does not meet the requirements as a signal-poor period, and when it is determined that the positioning accuracy of the quadruped robot is controllable according to the distribution of the signal-poor periods within the current preset time period of the quadruped robot, proceed to the next step;

[0009] S3: According to the walking stride and frequency of the four legs of the quadruped robot, use a speed fitting function to determine the speed information of the quadruped robot, and based on the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot, compensate for the course error and magnetic declination through the attitude angles to obtain the motion attitude information of the quadruped robot;

[0010] S4: Use the information of the previous moment of the quadruped robot and the motion attitude information and speed information of the current moment to perform navigation position calculation, calculate the position information of the current moment, and perform attitude control of the quadruped robot based on the position information and the compensation data within the current preset time period until positioning information with satisfactory quality is obtained, and then update the navigation position in real time.

[0011] A further technical solution lies in that the positioning information of the quadruped robot is determined according to the satellite positioning device of the quadruped robot.

[0012] A further technical solution lies in that the method for determining the signal quality of the positioning information is as follows:

[0013] Based on the positioning information, characteristic quantities of signal quality dimensions in different dimensions of the positioning data are determined, and the signal quality of the positioning information is determined through the weighted sum of the characteristic quantities of signal quality dimensions in different dimensions.

[0014] A further technical solution lies in that when the signal quality of the positioning information meets the requirements, the navigation position is updated based on the positioning information, and navigation management is performed based on the navigation position.

[0015] A further technical solution lies in that when the compensation amount of any one of the heading error or magnetic declination at the moment is greater than the preset compensation amount threshold, the moment is determined as the error compensation moment.

[0016] A further technical solution lies in that when the change situation of the position between the moment and the previous moment is not within the preset range, the moment is determined as the position change abnormal moment.

[0017] A further technical solution lies in that the attitude control of the quadruped robot is performed through a preset attitude control strategy, which specifically includes:

[0018] Controlling the quadruped robot to pause movement until the signal quality of the positioning information of the quadruped robot meets the requirements.

[0019] On the other hand, the present invention provides a low-cost combined navigation system for a quadruped robot in a complex environment, adopting the above-mentioned low-cost combined navigation method for a quadruped robot in a complex environment, which is characterized in that it specifically includes:

[0020] A signal quality evaluation module, a positioning accuracy evaluation module, a motion attitude acquisition module, and a navigation processing module;

[0021] The signal quality evaluation module is responsible for obtaining the positioning information of the quadruped robot in real time and judging whether the signal quality of the positioning information meets the requirements;

[0022] The positioning accuracy evaluation module is responsible for taking the time period when the signal quality does not meet the requirements as a poor signal time period, and determining whether the positioning accuracy of the quadruped robot is controllable according to the distribution of the poor signal time periods within the current preset time period of the quadruped robot;

[0023] The motion posture acquisition module is responsible for determining the speed information of the quadruped robot by using a speed fitting function according to the quadruped walking stride and frequency of the quadruped robot, and obtaining the motion posture information of the quadruped robot by compensating the heading error and the magnetic declination angle based on the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot;

[0024] The navigation processing module is responsible for calculating the navigation position using the information of the quadruped robot at the previous moment and the motion posture information and speed information at the current moment, calculating the position information at the current moment, and controlling the posture of the quadruped robot based on the position information and the compensation data within the current preset time period until the positioning information with quality that meets the requirements is obtained, and then updating the navigation position in real time.

[0025] The beneficial effects of the present invention are:

[0026] The quadruped robot navigation system fully utilizes the quadruped motion characteristics of the quadruped robot, low-cost satellite positioning equipment, and the quadruped robot's own inertial measurement unit. It can meet the needs of autonomous navigation of the quadruped robot and effectively solve the problems of high cost and complex system of existing quadruped robot navigation.

[0027] Other features and advantages will be described in the following description. The objects and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and drawings.

[0028] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0030] Figure 1 It is a flow chart of a low-cost integrated navigation method for a quadruped robot in a complex environment;

[0031] Figure 2 is a flow chart of a method for determining positioning accuracy of a quadruped robot;

[0032] Figure 3 is a flow chart of a method for determining velocity information of a quadruped robot;

[0033] Figure 4 is a flow chart of a method for determining posture control of a quadruped robot;

[0034] Figure 5 It is a framework diagram of a low-cost integrated navigation system for a quadruped robot in complex environments;

[0035] Figure 6 It is a carrier coordinate system diagram of a quadruped robot;

[0036] Figure 7 It is a diagram showing the relationship between the navigation position error and time of a quadruped robot. Specific implementation manners

[0037] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0038] (1) Currently existing technologies

[0039] Currently, the navigation of quadruped robots mainly uses lidar or depth cameras, and integrates the information output in real time by the inertial measurement unit (IMU) installed on the quadruped robot to perform SLAM navigation; when performing tasks in a large area, satellite positioning devices (GNSS) are often installed for combined navigation. Precise real-time positioning of the quadruped robot can be achieved through lidar / IMU / GNSS combined navigation or depth camera / IMU / GNSS combined navigation. At the same time, the precise position information of the quadruped robot can be transmitted to the control end in real time for display, so as to facilitate the control or monitoring by the control end. With the rapid development of the motion control technology of quadruped robots, the motion control of quadruped robots has gradually shifted from traditional model control (relying on the fusion information of lidar / IMU or depth camera / IMU) to motion control based on perceptionless reinforcement learning. Perceptionless reinforcement learning can effectively improve the motion control ability of quadruped robots and does not require relying on environment perception sensors, but high-precision navigation sensors need to be added for autonomous navigation.

[0040] (2) Defects of the existing technologies

[0041] Both lidar / IMU / GNSS combined navigation or depth camera / IMU / GNSS combined navigation can provide high-quality navigation information, but there are problems of relatively high sensor hardware costs and relatively large combined navigation software computing amounts; it is difficult for a quadruped robot to directly perform autonomous navigation based on perceptionless reinforcement learning motion control, and a high-precision navigation system needs to be installed, which directly increases the system cost and complexity. That is, the disadvantages of the existing technologies can be summarized as: 1) The existing lidar / IMU / GNSS combined navigation or depth camera / IMU / GNSS combined navigation has high costs and large software computing amounts; 2) For a quadruped robot based on perceptionless reinforcement learning motion control, new navigation devices need to be added, increasing the system cost and system complexity.

[0042] The present invention relates to a low-cost combined navigation system for quadruped robots in complex environments. This system realizes autonomous navigation in complex environments through the gait, stride frequency, inertial measurement unit, and satellite positioning device of the quadruped robot. The specific implementation sensor of the present invention is a WHEELTEC type GPS / BD dual-mode GNSS positioning module, and the rest of the equipment is a bionic quadruped robot, which includes a YIS320 type inertial measurement unit, a customized RK3588 motion control board, and a navigation information processing board. The software deployment environment is the Ubuntu18.04 operating system and the ROS Melodic robot operating system; the power supply of each device is provided by the quadruped robot battery pack after passing through a pressure plate conversion board.

[0043] Specifically, a low-cost combined navigation system for quadruped robots in complex environments has the following specific implementation steps:

[0044] Step 1: Establish the carrier coordinate system and navigation coordinate system of the quadruped robot. Using the motion control model of the quadruped robot and external speed measurement equipment, measure and record the actual motion speed of the quadruped robot, the stride and frequency of the quadruped walking of the quadruped robot, and use the least squares method to fit the relationship between the actual motion speed of the quadruped robot and the stride and frequency of the quadruped walking of the quadruped robot;

[0045] Step 2: Based on the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot, perform precise error compensation on the installation attitude angles, and compensate for the installation error angle and magnetic declination angle of the heading angle.

[0046] Step 3: When the quadruped robot works in a complex environment and the satellite positioning device cannot output positioning information in real time, use the information of the quadruped robot at the previous moment and the attitude and speed information at the current moment to perform navigation position inference and calculate the position information at the current moment;

[0047] Step 4: When the satellite positioning device of the quadruped robot can obtain high-quality positioning information, it can update the position deduced by dead reckoning in real time to correct the position error accumulated during navigation position inference;

[0048] Step 5: Analyze the errors of the navigation solution model and evaluate the navigation errors of the quadruped robot.

[0049] Step 6: Build and debug the software and hardware environment to ensure that each device works normally; develop navigation software based on the above process and navigation solution schematic diagram, and verify the navigation ability on the quadruped robot.

[0050] Embodiment 1

[0051] To solve the above problems, according to one aspect of the present invention, as Figure 1As shown, a low-cost combined navigation method for quadruped robots in complex environments is provided, characterized in that:

[0052] S1: Obtain the positioning information of the quadruped robot in real time, and when the signal quality of the positioning information does not meet the requirements, proceed to the next step;

[0053] Furthermore, the positioning information of the quadruped robot is determined according to the satellite positioning device of the quadruped robot.

[0054] Specifically, the method for determining the signal quality of the positioning information is as follows:

[0055] Based on the positioning information, determine the characteristic quantities of the signal quality dimensions in different dimensions of the positioning data, and determine the signal quality of the positioning information through the weighted sum of the characteristic quantities of the signal quality dimensions in different dimensions.

[0056] Furthermore, when the signal quality of the positioning information meets the requirements, update the navigation position based on the positioning information and perform navigation management based on the navigation position.

[0057] S2: Regard the time period when the signal quality does not meet the requirements as a signal-poor time period, and when it is determined that the positioning accuracy of the quadruped robot is controllable according to the distribution of the signal-poor time periods within the current preset time period of the quadruped robot, proceed to the next step;

[0058] Specifically, as Figure 2 shown, the method for determining the positioning accuracy of the quadruped robot is as follows:

[0059] Based on the distribution of the signal-poor time periods, obtain the number of signal-poor time periods of the quadruped robot within the preset time period;

[0060] Determine the positioning accuracy influence values of different signal-poor time periods through the duration of different signal-poor time periods and the signal quality at different moments of the signal-poor time periods;

[0061] Obtain the number of signal-poor time periods of the quadruped robot within the preset time period, and determine the positioning accuracy of the quadruped robot by combining the positioning accuracy influence values of different signal-poor time periods and the interval duration of different signal-poor time periods.

[0062] Furthermore, the value range of the positioning accuracy of the quadruped robot is between 0 and 1. When the positioning accuracy of the quadruped robot is not less than the preset accuracy, it is determined that the positioning accuracy of the quadruped robot is controllable.

[0063] Specifically, when the positioning accuracy of the quadruped robot is uncontrollable, the quadruped robot is controlled to pause until the signal quality of the positioning information of the quadruped robot meets the requirements.

[0064] In another embodiment, the method for determining the positioning accuracy of the quadruped robot is as follows:

[0065] Based on the distribution of the signal poor periods, obtain the cumulative duration of the signal poor periods of the quadruped robot within a preset time period, and determine whether the cumulative duration of the signal poor periods of the quadruped robot within the preset time period meets the requirements. If so, proceed to the next step; if not, determine that the positioning accuracy of the quadruped robot cannot meet the requirements.

[0066] Take the update moment of the nearest navigation position as the position update moment, and determine whether the interval duration between the position update moment and the current moment meets the requirements. If so, proceed to the next step; if not, determine that the positioning accuracy of the quadruped robot cannot meet the requirements.

[0067] Determine the current signal poor period according to the position update moment, and determine the positioning accuracy influence value of the current signal poor period according to the duration of the current signal poor period and the signal quality at different moments of the current signal poor period. Determine whether the positioning accuracy influence value of the current signal poor period meets the requirements. If so, proceed to the next step; if not, determine that the positioning accuracy of the quadruped robot cannot meet the requirements.

[0068] Determine the positioning accuracy influence values of different signal poor periods through the durations of different signal poor periods and the signal qualities at different moments of the signal poor periods, obtain the number of signal poor periods of the quadruped robot within a preset time period, and determine the positioning accuracy of the quadruped robot in combination with the positioning accuracy influence values of different signal poor periods and the interval durations of different signal poor periods.

[0069] S3 According to the quadruped walking stride and frequency of the quadruped robot, use a speed fitting function to determine the speed information of the quadruped robot, based on the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot, and compensate for the heading error and magnetic declination through the attitude angles to obtain the motion attitude information of the quadruped robot.

[0070] As Figure 6 shown, it is the carrier coordinate system diagram of the quadruped robot. Establish the carrier coordinate system of the quadruped robot (denoted as O b X b Y b Z b ), and the navigation coordinate system is the northeast celestial coordinate system (denoted as O n Xn Y n Z n ) Based on the motion characteristics of the quadruped robot, the walking stride of the quadruped robot is S i , and the frequency is F i , the leg-foot walking speed V of the quadruped robot bi , where i = 1, 2, 3, 4, that is, the motion amplitudes, frequencies, and speeds of the left, right, front, and rear leg-feet of the quadruped robot. The walking stride S of the quadruped robot i , and the frequency is F i can all be obtained in real time by the motion control system of the quadruped robot. Then the leg-foot walking speed V of the quadruped robot bi is

[0071]

[0072] To simplify the walking speed of the quadruped robot, the walking speed V of the quadruped robot b is

[0073]

[0074] According to the measured walking speed V of the quadruped robot b and the corresponding recorded leg-foot stride S of the quadruped robot i and frequency F i , using equations (1) and (2), a 3rd-order linear model and the least squares method are used to fit the relationship between the walking speed V of the quadruped robot b and S i , F i as

[0075] V b = f(S i , F i ) (3)

[0076] Place the quadruped robot on a horizontal ground environment and in a standard standing state. Error compensation is performed according to the attitude angle information output by the inertial measurement unit of the quadruped robot, so that when the quadruped robot is in a standard standing state on the horizontal ground, the pitch angle and roll angle real-time output angles are close to 0 degrees. At this time, the attitude error compensation of the pitch angle and roll angle of the quadruped robot is completed.

[0077] Use a high-precision GNSS differential positioning device to mark a straight line with a length of not less than 200 meters in an open horizontal environment as d. Control the quadruped robot to walk in a straight line according to the motion model, and record the minimum distance deviation between the walking straight line and the actual target point position as Δd. Check the magnetic declination corresponding to the longitude and latitude coordinates of the location Then, calculate the compensation angle of the heading angle of the quadruped robot as

[0078]

[0079] So far, the attitude angle error, heading angle error, and magnetic declination compensation of the quadruped robot have been completed.

[0080] Feature of Step 3: When the satellite navigation device of the quadruped robot cannot obtain effective longitude and latitude information, navigation calculation is required, and combined with Equation (3) and the pitch angle θ, roll angle γ, and heading angle output and compensated in real time by the inertial measurement unit of the quadruped robot Convert the velocity V in the body coordinate system b to the velocity V in the navigation coordinate system n as

[0081]

[0082] where, V E , V E , V u are the velocities of the quadruped robot in the east, north, and up directions in the navigation coordinate system respectively. is

[0083]

[0084] Integrating the velocity in the navigation coordinate system gives the longitude λ and latitude L at the current moment as

[0085]

[0086] where, λ 0 , L 0 are the longitude and latitude at the previous moment respectively, h is the altitude, R M , R N are the curvature radii of the meridian and prime vertical circles of the earth respectively, and are respectively

[0087]

[0088] where, a is the semi-major axis of the earth ellipsoid; e is the first eccentricity of the earth.

[0089] Combining the above Equations (3) to (10), the autonomous navigation of the quadruped robot when the satellite positioning signal fails in a complex environment can be completed.

[0090] Specifically, as Figure 3 shown, the method for determining the velocity information of the quadruped robot is:

[0091] Use a speed measurement device to measure the walking speed of the quadruped robot at different gaits and different gait frequencies, use the walking stride and frequency as the input of the speed fitting function, and use the speed measured by the speed measurement device as the output of the speed fitting function;

[0092] The parameters of the speed fitting function of the walking speed of the quadruped robot with respect to the walking stride and frequency of the quadruped robot are fitted by the least squares method, and the speed information of the quadruped robot walking is determined based on the speed fitting function and the walking stride and frequency of the quadruped robot.

[0093] Specifically, the specific steps for constructing the motion attitude information are as follows:

[0094] Use a high-precision GNSS differential positioning device to mark a straight-line length of no less than 200 meters in an open horizontal environment, control the quadruped robot to walk in a straight line according to the motion model, record the minimum distance deviation between the walking straight line and the actual target point position, and look up the compensation angles of the magnetic declination and heading angle corresponding to the latitude and longitude coordinates of the location.

[0095] According to the minimum distance deviation between the walking straight line and the actual target point position, look up the compensation angles of the magnetic declination and heading angle corresponding to the latitude and longitude coordinates of the location to perform the compensation of the heading error and magnetic declination to obtain the motion attitude information of the quadruped robot.

[0096] S4 uses the information of the quadruped robot at the previous moment, the motion attitude information and speed information at the current moment to perform navigation position calculation, calculates the position information at the current moment, and performs attitude control of the quadruped robot based on the position information and the compensation data within the current preset time period until the positioning information with qualified quality is obtained, and then updates the navigation position in real time.

[0097] When the satellite signal fails, step 3) can be used to complete the autonomous navigation of the quadruped robot, but there will be a phenomenon that the position error diverges with time during long-term navigation. To damp the problem of the navigation position error diverging with time, when the satellite positioning device of the quadruped robot can obtain the positioning position information, the positioning information output by the satellite positioning device is updated to the position solved in step 3) in time, which can effectively damp the divergence of the position error during the long-term navigation of the quadruped robot.

[0098] Feature of Step 5: By analyzing the above navigation solution algorithm and the errors of the selected sensors, the attitude and heading errors are mainly sensor errors. By referring to the technical manual of the YIS320 inertial measurement unit, it can be known that the roll angle error and pitch angle error of the quadruped robot navigation output are both about 0.05° (root mean square value, RMS), and the heading angle error is about 0.3° (RMS). The speed error of the quadruped robot navigation output is directly related to the gait measurement accuracy, step frequency measurement accuracy, fitting model error, terrain conditions, etc. This error needs to be verified in combination with the integrated navigation accuracy; the position error of the quadruped robot navigation output is the integrated navigation solution error and satellite positioning error. Based on the output accuracy of the WHEELTEC type GPS / BD dual-mode GNSS positioning module, when the positioning information output by the satellite positioning module is valid, the longitude and latitude errors of the quadruped robot can be controlled within 10 meters, and the satellite positioning error will not accumulate and diverge over time; therefore, the position error of the quadruped robot is mainly the integrated navigation solution error.

[0099] When the quadruped robot works in a complex environment and the satellite positioning module cannot work properly, the main error ΔS of calculating the navigation position of the quadruped robot is

[0100]

[0101] where ΔS λ , ΔS L are the position errors of the quadruped robot in the longitude and latitude directions respectively, ΔV b is the speed model error after fitting the gait and step frequency of the quadruped robot, V b is the walking speed of the quadruped robot, Δθ is the heading angle error after compensation of the output of the inertial measurement unit of the quadruped robot, and t is the walking time of the quadruped robot.

[0102] Set the speed error ΔV b of the quadruped robot to 0.05 m / s, the angle error Δθ of the inertial measurement unit of the quadruped robot to 0.3° (RMS), and the walking time t of the quadruped robot to 600 s. Simulate and calculate the walking speed V b of the quadruped robot at 1 m / s, 2 m / s, and 3 m / s respectively. The relationship between the navigation position error and time is as Figure 7 .

[0103] Feature of Step 6: Build the hardware of the quadruped robot navigation system, develop the navigation software according to Equations (1) to (10) and Figure 1 the navigation solution principle. After development, run it on the quadruped robot navigation board, and the quadruped robot can achieve autonomous navigation in a complex environment.

[0104] Specifically, as Figure 4 shown, the method for determining the attitude control of the quadruped robot is:

[0105] Determine the real-time position of the quadruped robot based on the position information, and determine the position accuracy of the quadruped robot based on the change in the real-time position and the position at the previous moment;

[0106] Obtain the compensation data of the heading error and magnetic declination at different moments within a preset time period, and determine the position positioning reliability of the quadruped robot at different moments in combination with the change in the position at different moments and the previous moment;

[0107] Correct the position accuracy of the quadruped robot through the position positioning reliability of the quadruped robot at different moments to obtain the comprehensive position reliability, and perform attitude control of the quadruped robot based on the comprehensive position reliability.

[0108] Further, performing attitude control of the quadruped robot based on the comprehensive position reliability specifically includes:

[0109] When the comprehensive position reliability meets the requirements, perform attitude control of the quadruped robot based on the traveling speed corresponding to the comprehensive position reliability;

[0110] When the comprehensive position reliability does not meet the requirements, control the quadruped robot to pause until the signal quality of the positioning information of the quadruped robot meets the requirements.

[0111] It can be understood that correcting the position accuracy of the quadruped robot through the position positioning reliability of the quadruped robot at different moments to obtain the comprehensive position reliability specifically includes:

[0112] Determine the basic weight values at different moments based on the duration from different moments to the current moment, and determine the weight sum of the position positioning reliability of the quadruped robot in combination with the position positioning reliability of the quadruped robot at different moments;

[0113] Perform normalization processing on the weight sum of the position positioning reliability to obtain a reliability correction amount, and determine the comprehensive position reliability through the sum of the reliability correction amount and the position accuracy of the quadruped robot.

[0114] In another embodiment, the method for determining the attitude control of the quadruped robot is:

[0115] Determine the real-time position of the quadruped robot based on the position information, and determine the position accuracy of the quadruped robot based on the change situation between the real-time position and the position at the previous moment. Judge whether the position accuracy of the quadruped robot meets the requirements. If so, proceed to the next step. If not, perform attitude control on the quadruped robot through a preset attitude control strategy;

[0116] Obtain the compensation data of the heading error and magnetic declination at different moments within a preset time period, and determine the error compensation moment according to the compensation data. Judge whether the number of the deviation compensation moments meets the requirements. If so, proceed to the next step. If not, perform attitude control on the quadruped robot through a preset attitude control strategy;

[0117] Determine the abnormal position change moment of the quadruped robot according to the change situation between different moments and the position at the previous moment. Judge whether the number of the abnormal position change moments of the quadruped robot meets the requirements. If so, proceed to the next step. If not, perform attitude control on the quadruped robot through a preset attitude control strategy;

[0118] Obtain the compensation data of the heading error and magnetic declination at different moments within a preset time period, and determine the position positioning reliability of the quadruped robot at different moments in combination with the change situation between different moments and the position at the previous moment. Judge whether the number of the moments when the position positioning reliability does not meet the requirements meets the requirements. If so, proceed to the next step. If not, perform attitude control on the quadruped robot through a preset attitude control strategy;

[0119] Correct the position accuracy of the quadruped robot through the position positioning reliability of the quadruped robot at different moments to obtain the comprehensive position reliability, and perform attitude control on the quadruped robot based on the comprehensive position reliability.

[0120] Furthermore, when the compensation amount of any one of the heading error or magnetic declination at the moment is greater than the preset compensation amount threshold, then determine that the moment is the error compensation moment.

[0121] Specifically, when the change situation between the position at the moment and the position at the previous moment is not within the preset range, then determine that the moment is the abnormal position change moment.

[0122] It should be noted that performing attitude control on the quadruped robot through a preset attitude control strategy specifically includes:

[0123] Control the quadruped robot to pause movement until the signal quality of the positioning information of the quadruped robot meets the requirements.

[0124] Embodiment 2

[0125] On the other hand, Figure 5 As shown, the present invention provides a low-cost integrated navigation system for a quadruped robot in a complex environment, which adopts the above-mentioned low-cost integrated navigation method for a quadruped robot in a complex environment, and is characterized in that it specifically includes:

[0126] Signal quality assessment module, positioning accuracy assessment module, motion posture acquisition module, navigation processing module;

[0127] The signal quality evaluation module is responsible for acquiring the positioning information of the quadruped robot in real time and determining whether the signal quality of the positioning information meets the requirements;

[0128] The positioning accuracy evaluation module is responsible for taking the time period when the signal quality does not meet the requirements as the signal poor time period, and determining whether the positioning accuracy of the quadruped robot is controllable according to the distribution of the signal poor time period within the current preset time period of the quadruped robot;

[0129] The motion posture acquisition module is responsible for determining the speed information of the quadruped robot by using a speed fitting function according to the quadruped walking stride and frequency of the quadruped robot, and obtaining the motion posture information of the quadruped robot by compensating the heading error and the magnetic declination angle based on the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot;

[0130] The navigation processing module is responsible for calculating the navigation position using the information of the quadruped robot at the previous moment and the motion posture information and speed information at the current moment, calculating the position information at the current moment, and controlling the posture of the quadruped robot based on the position information and the compensation data within the current preset time period until the positioning information with quality that meets the requirements is obtained, and then updating the navigation position in real time.

[0131] Through the above embodiments, the present invention achieves the following beneficial effects:

[0132] The quadruped robot navigation system fully utilizes the quadruped motion characteristics of the quadruped robot, low-cost satellite positioning equipment, and the quadruped robot's own inertial measurement unit. It can meet the needs of autonomous navigation of the quadruped robot and effectively solve the problems of high cost and complex system of existing quadruped robot navigation.

[0133] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0134] The above description has been made of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0135] The foregoing is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. A low-cost integrated navigation method for a quadruped robot in a complex environment, characterized by: Acquire the positioning information of the quadruped robot in real time, and proceed to the next step when the signal quality of the positioning information does not meet the requirements; The time period when the signal quality does not meet the requirements is regarded as a poor signal period, and when it is determined that the positioning accuracy of the quadruped robot is controllable according to the distribution of the poor signal periods within the current preset time period of the quadruped robot, the next step is entered; According to the walking stride and frequency of the quadruped robot, the speed information of the quadruped robot is determined by using the speed fitting function. The motion posture information of the quadruped robot is obtained by compensating the heading error and magnetic declination through the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot. The navigation position is estimated using the information of the quadruped robot at the previous moment and the motion posture information and speed information at the current moment, and the position information at the current moment is calculated. The posture of the quadruped robot is controlled based on the position information and the compensation data within the current preset time period until the positioning information with quality that meets the requirements is obtained, and the navigation position is updated in real time.

2. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: The positioning information of the quadruped robot is determined based on the satellite positioning device of the quadruped robot.

3. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: The method for determining the signal quality of the positioning information is: Based on the positioning information, feature quantities of signal quality dimensions of the positioning data in different dimensions are determined, and the signal quality of the positioning information is determined by summing the weights of the feature quantities of signal quality dimensions in different dimensions.

4. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: When the signal quality of the positioning information meets the requirement, the navigation position is updated based on the positioning information, and navigation management is performed based on the navigation position.

5. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: The method for determining the positioning accuracy of the quadruped robot is: Obtaining the number of poor signal periods of the quadruped robot within a preset time period based on the distribution of the poor signal periods; Determine the positioning accuracy impact values ​​of different periods of poor signal by the duration of different periods of poor signal and the signal quality at different times of the period of poor signal; The number of periods of poor signal of the quadruped robot within a preset time period is obtained, and the positioning accuracy of the quadruped robot is determined in combination with the positioning accuracy impact values ​​of different periods of poor signal and the interval durations of different periods of poor signal.

6. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 5, characterized in that: When the positioning accuracy of the quadruped robot is uncontrollable, the quadruped robot is controlled to suspend movement until the signal quality of the positioning information of the quadruped robot meets the requirements.

7. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: The method for determining the positioning accuracy of the quadruped robot is: Based on the distribution of the signal-poor time periods, the cumulative duration of the signal-poor time periods of the quadruped robot within a preset time period is obtained, and it is determined whether the cumulative duration of the signal-poor time periods of the quadruped robot within the preset time period meets the requirement, if so, proceeding to the next step, if not, determining that the positioning accuracy of the quadruped robot cannot meet the requirement; The update time of the nearest navigation position is used as the position update time, and whether the interval between the position update time and the current time meets the requirement is determined. If so, the next step is entered; if not, it is determined that the positioning accuracy of the quadruped robot cannot meet the requirement; Determine the current signal poor period according to the position update time, and determine the positioning accuracy impact value of the current signal poor period according to the duration of the current signal poor period and the signal quality at different times of the current signal poor period, and judge whether the positioning accuracy impact value of the current signal poor period meets the requirements. If so, proceed to the next step; if not, determine that the positioning accuracy of the quadruped robot cannot meet the requirements; The positioning accuracy impact values ​​of different periods of poor signal are determined by the duration of different periods of poor signal and the signal quality at different times during the periods of poor signal, the number of periods of poor signal of the quadruped robot within a preset time period is obtained, and the positioning accuracy of the quadruped robot is determined in combination with the positioning accuracy impact values ​​of different periods of poor signal and the interval durations of different periods of poor signal.

8. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: The method for determining the speed information of the quadruped robot is: A speed measuring device is used to measure the walking speed of the quadruped robot under different gaits and different gait frequencies, and the walking stride and frequency are used as inputs of a speed fitting function, and the speed measured by the speed measuring device is used as output of the speed fitting function; The least square method is used to fit the parameters of the speed fitting function of the walking speed of the quadruped robot and the walking stride and frequency of the quadruped robot, and the walking speed information of the quadruped robot is determined based on the speed fitting function and the walking stride and frequency of the quadruped robot.

9. The low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in claim 1, characterized in that: The specific steps of constructing the motion posture information are: Use high-precision GNSS differential positioning equipment to mark a straight line length of no less than 200 meters in an open horizontal environment, control the quadruped robot to walk in a straight line according to the motion model, record the minimum distance deviation between the walking straight line and the actual target point position, and check the compensation angle of the magnetic declination and heading angle corresponding to the longitude and latitude coordinates of the location; According to the minimum distance deviation between the walking straight line and the actual target point position, the compensation angles of the magnetic declination and heading angle corresponding to the latitude and longitude coordinates of the position are consulted to compensate for the heading error and the magnetic declination to obtain the motion posture information of the quadruped robot.

10. A low-cost integrated navigation system for a quadruped robot in a complex environment, using a low-cost integrated navigation method for a quadruped robot in a complex environment as claimed in any one of claims 1 to 9, characterized in that: Specifically include: Signal quality assessment module, positioning accuracy assessment module, motion posture acquisition module, navigation processing module; The signal quality evaluation module is responsible for acquiring the positioning information of the quadruped robot in real time and determining whether the signal quality of the positioning information meets the requirements; The positioning accuracy evaluation module is responsible for taking the time period when the signal quality does not meet the requirements as the signal poor time period, and determining whether the positioning accuracy of the quadruped robot is controllable according to the distribution of the signal poor time period within the current preset time period of the quadruped robot; The motion posture acquisition module is responsible for determining the speed information of the quadruped robot by using a speed fitting function according to the quadruped walking stride and frequency of the quadruped robot, and obtaining the motion posture information of the quadruped robot by compensating the heading error and the magnetic declination angle based on the three attitude angles output in real time by the inertial measurement unit installed on the quadruped robot; The navigation processing module is responsible for calculating the navigation position using the information of the quadruped robot at the previous moment and the motion posture information and speed information at the current moment, calculating the position information at the current moment, and controlling the posture of the quadruped robot based on the position information and the compensation data within the current preset time period until the positioning information with quality that meets the requirements is obtained, and then updating the navigation position in real time.

Images