Autonomous integrity assessment method for multi-source fusion positioning in complex environments

Abstract

The present invention discloses an autonomous integrity assessment method for multi-source fusion positioning in a complex environment, comprising the following steps: A. constructing a GNSS PPP-RTK / INS tightly combined positioning method with a priori zero-value deviation constraints; B. constructing a zero-value deviation integrity discrimination equation; C. constructing a position and attitude joint lateral integrity assessment method, and then, based on a sensor attitude anomaly flag, adopting a PPP-RTK / INS / laser SLAM multi-source fusion estimation based on a graph optimization fusion positioning method to obtain the fused position and attitude of the intelligent rescue equipment after GNSS PPP-RTK / INS / laser SLAM; D. using a zero-value deviation-based correctness assessment equation to discriminate the correctness of the current zero-value deviation update, and updating the inertial navigation acceleration zero-value deviation and angular velocity zero-value deviation based on the zero-value deviation update correctness flag. The present invention fully utilizes the characteristics of the short-term stability of the inertial navigation zero-value deviation and the convergence of the postures of the carrier positioning sensors, ultimately achieving continuous, reliable and high-precision positioning of the intelligent rescue equipment.

Description

Technical Field

[0001] The present invention belongs to the field of positioning technology and provides an autonomous integrity assessment method for multi-source fusion positioning in complex environments. The method is suitable for high-precision fusion positioning in the integrity assessment of intelligent rescue platforms under complex environmental conditions such as satellite occlusion and lidar failure. Background Art

[0002] Fused positioning based on GNSS, lidar, and inertial navigation provides intelligent rescue equipment with high-precision positioning information in most scenarios. However, in complex environments where multi-source fused positioning is inefficient or even fails due to satellite obstruction and lidar failure, how to improve the intelligent rescue equipment's ability to output continuous high-precision and highly reliable positioning information is an important factor in expanding the intelligent application of emergency rescue equipment and realizing autonomous task handling by intelligent equipment.

[0003] Multi-sensor fusion positioning technology based on GNSS, INS and lidar has become an important technical approach for obtaining positioning information for intelligent rescue equipment. However, in complex environmental conditions such as satellite occlusion and lidar failure, there are still defects such as reduced efficiency or even failure of fusion positioning, which are mainly reflected in the following aspects: ① GNSS high-precision positioning is heavily dependent on high-quality satellite pseudorange and carrier phase measurement data. Poor observation quality in complex environments leads to GNSS positioning anomalies, affecting the robustness of the multi-source fusion positioning system; ② The current lidar SLAM-based multi-sensor fusion positioning method is mainly aimed at feature-rich urban environments or indoor environments, and is not suitable for extreme environments such as tunnels and dense forests, resulting in the risk of re-initialization of the multi-source fusion positioning system; ③ The multi-source fusion high-precision positioning system framework based on posterior statistics relies heavily on the posterior statistics output by the fusion estimation system, and is unable to identify the anomaly of a sensor in real time and isolate it before fusion, which easily causes the multi-source fusion estimation system to be contaminated, resulting in systematic deviations in the subsequent multi-source fusion positioning results. Summary of the Invention

[0004] The problem to be solved by the present invention is that in complex environments such as satellite signal obstruction and lidar failure, some sensor failures lead to poor robustness of multi-source fusion positioning or even failure of fusion positioning. A method for autonomous integrity assessment of multi-source fusion positioning in complex environments is provided.

[0005] To solve the above technical problems, the present invention adopts a technical solution: a method for autonomous integrity assessment of multi-source fusion positioning in complex environments, comprising the following steps:

[0006] A. A GNSS PPP-RTK / INS tightly coupled positioning method with prior zero-bias constraints is constructed based on the previous moment's acceleration zero-bias and angular velocity zero-bias of the intelligent rescue equipment and combined with the equipment's current GNSS pseudorange, carrier phase observation data, and inertial navigation measurement data to estimate the equipment's current position, attitude, velocity, angular velocity zero-bias, and acceleration zero-bias.

[0007] B. The current position, velocity, angular velocity zero deviation, and acceleration zero deviation of the equipment obtained by the GNSS PPP-RTK / INS tight integration positioning method are combined with the angular velocity zero deviation and acceleration zero deviation at the previous moment to construct a zero deviation integrity discriminant equation to determine whether the GNSS PPP-RTK / INS estimation status is normal and identify abnormal conditions in the fusion solution;

[0008] C. Obtain the current position and attitude matrix based on the current INS raw acceleration information and angular velocity measurement information and the previous position, attitude, acceleration zero deviation, and angular velocity zero deviation information. Then, combine the GNSS PPP-RTK / INS fused position and attitude matrix and the current position and attitude matrix estimated by the laser SLAM of the intelligent rescue equipment to construct a joint position and attitude lateral integrity assessment method to identify abnormal state information of each sensor position and attitude. Then, based on the sensor attitude abnormality identification, a graph-optimized fusion positioning method is used to fuse the GNSS PPP-RTK / INS / laser SLAM fused position and attitude of the intelligent rescue equipment.

[0009] D. Based on the current position and attitude information of the intelligent rescue equipment GNSS PPP-RTK / INS / laser SLAM fusion, combined with the angular velocity measurement information of the current INS raw acceleration information, the current inertial navigation acceleration zero-value bias and angular velocity zero-value bias information are estimated using Kalman filter estimation technology. At the same time, combined with the inertial navigation acceleration zero-value bias and angular velocity zero-value bias information at historical moments, the correctness of the current zero-value bias update is determined based on the zero-value bias correctness evaluation equation. The inertial navigation acceleration zero-value bias and angular velocity zero-value bias are updated based on the zero-value bias update correctness flag.

[0010] Furthermore, the system state equation of the GNSS PPP-RTK / INS tight integration positioning method in step A is:

[0011]

[0012]

[0013] Among them, F represents the system state transfer matrix, G represents the system noise driving matrix, w represents the system noise vector, X and is the system state vector and its derivative, δr represents the position error, δt r Indicates the receiver clock deviation, represents the carrier phase integer deviation, δv represents the velocity error, b g Indicates the angular velocity zero value deviation, b a Indicates the acceleration zero value deviation, represents the misalignment angle error, and N represents the carrier phase ambiguity parameter;

[0014] The observation equation of the GNSS PPP-RTK / INS tight combination positioning method is:

[0015]

[0016]

[0017]

[0018] in, and They represent the pseudorange and carrier phase measurements of the frequency f of the GNSS satellite i observed by the intelligent rescue equipment, and represent the pseudorange noise and carrier phase noise of the f frequency of the GNSS satellite i observed by the intelligent rescue equipment, Represents the geometric distance between the intelligent rescue equipment and GNSS satellite i, I i and T i Respectively represent the ionospheric and tropospheric delay information of GNSS satellite i observed by the intelligent rescue equipment, δt i and δt represent the satellite clock error of GNSS satellite i and the receiver clock error of intelligent rescue equipment, respectively, λ f and represent the carrier phase wavelength and the carrier phase integer ambiguity of GNSS satellite i, respectively, represents the pseudorange code phase deviation of GNSS i satellite frequency f, represents the integer phase bias of GNSS i satellite, r0 and r s They represent the initial approximate position of the intelligent rescue equipment and the position vector of the GNSS satellite respectively.

[0019] Furthermore, step A also includes the step of determining whether the GNSS positioning technology is effective, which includes: constructing a GNSS effectiveness judgment equation based on the number of GNSS satellites of the intelligent rescue equipment at the current moment, the quality of pseudo-range carrier observations, and the age information of PPP-RTK differential enhancement data;

[0020] The GNSS validity judgment equation is:

[0021]

[0022] Among them, Idx GNSS Indicates whether GNSS is available, Nc and N s They represent the number of GNSS satellites that the intelligent rescue equipment observes complete pseudorange and carrier phase and their discrimination thresholds, ερ and δρ represent the noise level of GNSS pseudorange and its discrimination threshold, respectively. and They represent the noise level of the GNSS carrier phase and its discrimination threshold, respectively. δT and Ts represent the data age of the PPP-RTK differential enhancement information and its discrimination threshold, respectively.

[0023] Furthermore, the integrity judgment equation of the zero-value deviation in step B is:

[0024]

[0025] Among them, GNSS PPP-RTK / INS is the fusion solution anomaly indicator. and They represent the acceleration zero deviation stability threshold and the angular velocity zero deviation stability threshold, respectively, and b g Indicates the angular velocity zero value deviation, b a Indicates the acceleration zero value deviation, and They represent the virtual observation values ​​of the acceleration zero deviation and angular velocity zero deviation at the previous moment respectively;

[0026]

[0027]

[0028] in, and They represent the constraint matrices of the virtual observation equations for the zero-value deviation of acceleration and the zero-value deviation of angular velocity at the previous moment, and They represent the acceleration zero-value deviation measurement error and angular velocity zero-value deviation measurement error at the previous moment respectively.

[0029] Furthermore, the position and attitude combined lateral integrity assessment method in step C includes the following steps:

[0030] C.1 Identify abnormal status information of each sensor position and posture:

[0031]

[0032] Among them, Idvi is the abnormal position and posture mark of each sensor, r med and Rs. med The position and attitude rotation matrices are obtained based on median filtering at the current moment, r i and Rs. i The position and attitude rotation matrices obtained by each sensor, i can be PPP_RTK_INS or INS or SLAM;

[0033]

[0034]

[0035]

[0036] Among them, Median is the median filter function, ε r and ε Rs Respectively expressed as the position deviation judgment threshold and the attitude rotation matrix deviation judgment threshold

[0037] C.2 Using the above sensor position and posture abnormality identifier Idv i Based on, select the pose identifier Idv i The position and attitude matrix of each sensor with a value of 1 is used as the observation data, and the graph optimization fusion positioning method is adopted to obtain the position and attitude of the intelligent rescue equipment after GNSS PPP-RTK / INS / laser SLAM fusion.

[0038] Furthermore, in step D, the zero-value deviation correctness evaluation equation is:

[0039]

[0040] Among them, Idx bias Update the correctness flag for zero-value deviations.

[0041] Furthermore, in step D, updating the correctness flag according to the zero-value deviation, and updating the inertial navigation acceleration zero-value deviation and the angular velocity zero-value deviation include the following steps:

[0042]

[0043]

[0044] The beneficial effects of the present invention are: focusing on the frequent failure problems of GNSS PPP-RTK, lidar SLAM, etc. in complex environments, making full use of the characteristics of the short-term stability of inertial navigation zero-value deviation and the convergence of the postures of various positioning sensors of the carrier, proposing a GNSS PPP-RTK / INS and lidar layered distributed multi-sensor autonomous integrity assessment system with inertial navigation zero-value deviation and posture as the medium, designing a GNSS validity judgment equation with GNSS data quality constraints, a sensor anomaly recognition model based on zero-value deviation, and a position and posture joint lateral integrity assessment method, and finally realizing continuous and reliable high-precision positioning of intelligent rescue equipment.

[0045] The present invention will be described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of the autonomous integrity assessment method for multi-source fusion positioning in complex environments of the present invention. DETAILED DESCRIPTION

[0047] The method is based on sensors including GNSS, LiDAR, and an inertial measurement unit. The GNSS sensor collects carrier phase and pseudorange measurement data of the intelligent rescue equipment, the inertial navigation measurement unit collects angular velocity and acceleration information of the intelligent rescue equipment, and the LiDAR measures point cloud data around the intelligent rescue equipment, and outputs the position and posture information of the emergency rescue equipment based on laser SLAM.

[0048] Referring to the accompanying drawings, the method of the present invention includes the following steps.

[0049] Step A constructs a GNSS validity judgment equation based on the number of GNSS satellites of the intelligent rescue equipment at the current moment, the quality of pseudo-range carrier observations, and the age information of PPP-RTK differential enhancement data to identify whether the GNSS positioning technology method is effective.

[0050] The GNSS validity judgment equation is constructed as follows:

[0051]

[0052] Among them, Idx GNSS Indicates whether GNSS is available, Nc and N s They represent the number of GNSS satellites that the intelligent rescue equipment observes complete pseudorange and carrier phase and their discrimination thresholds, ερ and δρ represent the noise level of GNSS pseudorange and its discrimination threshold, respectively. and They represent the noise level of the GNSS carrier phase and its discrimination threshold, respectively. δT and Ts represent the data age of the PPP-RTK differential enhancement information and its discrimination threshold, respectively.

[0053] Under the condition that GNSS is available, based on the zero-value deviation of acceleration and angular velocity at the previous moment as constraints, combined with the current GNSS pseudorange, carrier phase observation data and inertial navigation measurement data of intelligent rescue equipment, a GNSS PPP-RTK / INS tight combination positioning method with prior zero-value deviation constraints is constructed to estimate the equipment position, attitude, velocity, zero-value deviation of angular velocity and zero-value deviation of acceleration at the current moment.

[0054] The GNSS PPP-RTK / INS tight integration positioning method includes:

[0055] The system state equation of the GNSS PPP-RTK / INS combined positioning method is:

[0056]

[0057]

[0058] Among them, F represents the system state transfer matrix, G represents the system noise driving matrix, w represents the system noise vector, X and is the system state vector and its derivative, δr represents the position error, δt r Indicates the receiver clock deviation, represents the carrier phase integer deviation, δv represents the velocity error, b g Indicates the angular velocity zero value deviation, b a Indicates the acceleration zero value deviation, represents the misalignment angle error, and N represents the carrier phase ambiguity parameter.

[0059] The observation equation of the GNSS PPP-RTK / INS tight combination method with prior zero bias constraint is:

[0060]

[0061]

[0062]

[0063] in, and They represent the pseudorange and carrier phase measurements of the frequency f of the GNSS satellite i observed by the intelligent rescue equipment, and represent the pseudorange noise and carrier phase noise of the f frequency of the GNSS satellite i observed by the intelligent rescue equipment, Represents the geometric distance between the intelligent rescue equipment and GNSS satellite i, I i and T iRespectively represent the ionospheric and tropospheric delay information of GNSS satellite i observed by the intelligent rescue equipment, δt i and δt represent the satellite clock error of GNSS satellite i and the receiver clock error of intelligent rescue equipment, respectively, λ f and represent the carrier phase wavelength and the carrier phase integer ambiguity of GNSS satellite i, respectively, represents the pseudorange code phase deviation of GNSS i satellite frequency f, represents the integer phase bias of GNSS i satellite, r0 and r s They represent the initial approximate position of the intelligent rescue equipment and the position vector of the GNSS satellite respectively.

[0064] Step B: Based on the GNSS PPP-RTK / INS combined positioning of the intelligent rescue equipment, the current position state, velocity, angular velocity zero deviation, and acceleration zero deviation are obtained. Combined with historical information such as the angular velocity zero deviation and acceleration zero deviation at the previous moment, a zero deviation integrity discriminant equation is constructed to determine whether the GNSS PPP-RTK / INS estimated state is normal and identify abnormal states of the fusion solution.

[0065] Based on the GNSS PPP-RTK / INS combined positioning of intelligent rescue equipment, the current position state r is obtained PPP_PTK_INS , speed v PPP_PTK_INS , angular velocity zero value deviation b g , acceleration zero value deviation b a , combined with the zero-value deviation of the angular velocity at the previous moment Acceleration zero value deviation information To judge the GNSS PPP-RTK / INS estimation status and identify abnormal conditions of the fusion solution, the constructed zero-value deviation integrity judgment equation is as follows:

[0066]

[0067] Among them, GNSS PPP-RTK / INS fusion solution anomaly identification, and They represent the acceleration zero-value deviation stability threshold and the angular velocity zero-value deviation stability threshold respectively.

[0068] The virtual equation constructed based on the prior zero-value deviation is:

[0069]

[0070]

[0071] in, and They represent the pseudorange and carrier phase measurements of the frequency f of the GNSS satellite i observed by the intelligent rescue equipment, and represent the pseudorange noise and carrier phase noise of the f frequency of the GNSS satellite i observed by the intelligent rescue equipment, Represents the geometric distance between the intelligent rescue equipment and GNSS satellite i, I i and T i Respectively represent the ionospheric and tropospheric delay information of GNSS satellite i observed by the intelligent rescue equipment, δt i and δt represent the satellite clock error of GNSS satellite i and the receiver clock error of intelligent rescue equipment, respectively.↓ f and represent the carrier phase wavelength and the carrier phase integer ambiguity of GNSS satellite i, respectively, represents the pseudorange code phase deviation of GNSS i satellite frequency f, represents the integer phase bias of GNSS i satellite, r0 and r s Represent the initial approximate position of the intelligent rescue equipment and the position vector of the GNSS satellite, and They represent the virtual observation values ​​of the acceleration zero deviation and angular velocity zero deviation at the previous moment, and They represent the constraint matrices of the virtual observation equations for the zero-value deviation of acceleration and the zero-value deviation of angular velocity at the previous moment, and They represent the acceleration zero-value deviation measurement error and angular velocity zero-value deviation measurement error at the previous moment respectively.

[0072] Step C: Based on the original acceleration information and angular velocity measurement information of the INS at the current moment, combined with the position, attitude, acceleration zero-value deviation and angular velocity zero-value deviation information at the previous moment, the current position and attitude matrix is ​​obtained by integration. Combined with the GNSS PPP-RTK / INS fused position and attitude matrix and the current position and attitude matrix estimated by the intelligent rescue equipment laser SLAM, a joint lateral integrity assessment method for position and attitude is constructed to identify the abnormal state information of the position and attitude of each sensor. Then, based on the sensor attitude abnormality identification, a graph-optimized fusion positioning method is used to obtain the fused position and attitude of the intelligent rescue equipment GNSS PPP-RTK / INS / laser SLAM.

[0073] According to the current moment INS original acceleration information, angular velocity measurement information, combined with the previous moment position, attitude, acceleration zero value deviation and angular velocity zero value deviation information, the current moment position r is obtained by integration INS and the posture matrix Rs INS, combined with the GNSS PPP-RTK / INS fusion position r PPP_PTK_INS and the attitude rotation matrix Rs PPP_PTK_INS , and the current position r estimated by the laser SLAM of the intelligent rescue equipment SLAM , attitude rotation matrix Rs SLAM , identify the abnormal status information of each sensor position and attitude, and construct the joint lateral integrity assessment method of position and attitude as follows:

[0074]

[0075]

[0076] Among them, r med and Rs. med The position and attitude rotation matrices are obtained based on median filtering at the current moment, Median is the median filter function, ε r and ε Rs They are respectively expressed as the position deviation judgment threshold and the attitude rotation matrix deviation judgment threshold.

[0077]

[0078] Among them, Idv i is the abnormal position and posture mark of each sensor, r i and Rs. i The position and attitude rotation matrices obtained by each sensor. The value of i is PPP_RTK_INS or INS or SLAM.

[0079] The above sensor position and posture abnormality identification Idv i Based on, select the pose identifier Idv i The position and attitude matrix of each sensor with a value of 1 are used as observation data. The graph optimization fusion positioning method is used to obtain the fusion position r of the intelligent rescue equipment GNSS PPP-RTK / INS / laser SLAM. com With gesture Rs com .

[0080] Step D: Based on the current position and attitude information of the intelligent rescue equipment GNSS PPP-RTK / INS / laser SLAM fusion, combined with the current INS original acceleration information and angular velocity measurement information, based on the Kalman filter estimation technology, estimate the current inertial navigation acceleration zero value deviation and angular velocity zero value deviation information. At the same time, combined with the historical inertial navigation acceleration zero value deviation and angular velocity zero value deviation information, based on the zero value deviation correctness evaluation equation, determine the correctness of the current zero value deviation update, and update the inertial navigation acceleration zero value deviation and angular velocity zero value deviation according to the zero value deviation update correctness flag.

[0081] Current position r based on GNSS PPP-RTK / INS / laser SLAM fusion of intelligent rescue equipment com and the attitude rotation matrix Rs com , combined with the current INS original acceleration information and angular velocity measurement information, based on the Kalman filter estimation technology, the current inertial navigation acceleration zero value deviation is estimated Angular velocity zero value deviation information Combined with the zero value deviation of inertial navigation acceleration at the previous moment Angular velocity zero value deviation information To determine the correctness of the current zero-value deviation update, the constructed zero-value deviation correctness evaluation equation is implemented as follows:

[0082]

[0083] Among them, Idx bias Update the correctness flag for zero-value deviations.

[0084] Update the correctness identifier Idx based on the zero-value deviation bias , update the inertial navigation acceleration zero value deviation b a and the angular velocity zero deviation b g , the update process is as follows:

[0085]

[0086]

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solutions of the present invention. They should all be included in the scope of the technical solutions claimed for protection by the present invention.

Claims

1. A method for autonomous integrity assessment of multi-source fusion positioning in complex environments, characterized by: The following steps are involved: A. A GNSS PPP-RTK / INS tightly coupled positioning method with prior zero-bias constraints is constructed based on the previous moment's acceleration zero-bias and angular velocity zero-bias of the intelligent rescue equipment and combined with the equipment's current GNSS pseudorange, carrier phase observation data, and inertial navigation measurement data to estimate the equipment's current position, attitude, velocity, angular velocity zero-bias, and acceleration zero-bias. B. The current position, velocity, angular velocity zero deviation, and acceleration zero deviation of the equipment obtained by the GNSS PPP-RTK / INS tight integration positioning method are combined with the angular velocity zero deviation and acceleration zero deviation at the previous moment to construct a zero deviation integrity discriminant equation to determine whether the GNSS PPP-RTK / INS estimation status is normal and identify abnormal conditions in the fusion solution; C. Obtain the current position and attitude matrix based on the current INS raw acceleration information and angular velocity measurement information and the previous position, attitude, acceleration zero-value deviation, and angular velocity zero-value deviation information. Then, combine the GNSSPPP-RTK / INS fused position and attitude matrix and the current position and attitude matrix estimated by the laser SLAM of the intelligent rescue equipment to construct a joint lateral integrity assessment method for position and attitude. Identify the abnormal state information of the position and attitude of each sensor. Then, based on the sensor attitude abnormality identification, adopt the PPP-RTK / INS / laser SLAM multi-source fusion estimation based on the graph optimization fusion positioning method to obtain the fused position and attitude of the intelligent rescue equipment after GNSS PPP-RTK / INS / laser SLAM. D. Based on the current position and attitude information of the intelligent rescue equipment GNSS PPP-RTK / INS / laser SLAM fusion, combined with the angular velocity measurement information of the current INS raw acceleration information, the current inertial navigation acceleration zero-value bias and angular velocity zero-value bias information are estimated using Kalman filter estimation technology. At the same time, combined with the inertial navigation acceleration zero-value bias and angular velocity zero-value bias information at historical moments, the correctness of the current zero-value bias update is determined using a correctness evaluation equation based on zero-value bias. The correctness flag is updated based on the zero-value bias, and the inertial navigation acceleration zero-value bias and angular velocity zero-value bias are updated.

2. The method for autonomous integrity assessment of multi-source fusion positioning in complex environments according to claim 1 is characterized in that: The system state equation of the GNSS PPP-RTK / INS tight integration positioning method in step A is: Among them, F represents the system state transfer matrix, G represents the system noise driving matrix, w represents the system noise vector, X and is the system state vector and its derivative, δr represents the position error, δt r Indicates the receiver clock deviation, represents the carrier phase integer deviation, δv represents the velocity error, b g Indicates the angular velocity zero value deviation, b a Indicates the acceleration zero value deviation, represents the misalignment angle error, and N represents the carrier phase ambiguity parameter; The observation equation of the GNSS PPP-RTK / INS tight combination positioning method is: in, and They represent the pseudorange and carrier phase measurements of the frequency f of the GNSS satellite i observed by the intelligent rescue equipment, and represent the pseudorange noise and carrier phase noise of the f frequency of the GNSS satellite i observed by the intelligent rescue equipment, Represents the geometric distance between the intelligent rescue equipment and GNSS satellite i, I i and T i Respectively represent the ionospheric and tropospheric delay information of GNSS satellite i observed by the intelligent rescue equipment, δt i and δt represent the satellite clock error of GNSS satellite i and the receiver clock error of intelligent rescue equipment, respectively, λ f and represent the carrier phase wavelength and the carrier phase integer ambiguity of GNSS satellite i, respectively, represents the pseudorange code phase deviation of the GNSSi satellite frequency f, Represents the integer phase bias of the GNSSi satellite, r0 and r s They represent the initial approximate position of the intelligent rescue equipment and the position vector of the GNSS satellite respectively.

3. The method for autonomous integrity assessment of multi-source fusion positioning in complex environments according to claim 1 is characterized in that: Step A also includes the step of determining whether the GNSS positioning technology is effective, which includes: constructing a GNSS effectiveness judgment equation based on the number of GNSS satellites of the intelligent rescue equipment at the current moment, the quality of pseudo-range carrier observations, and the age information of PPP-RTK differential enhancement data; The GNSS validity judgment equation is: Among them, Idx GNSS Indicates whether GNSS is available, Nc and N s They represent the number of GNSS satellites that the intelligent rescue equipment observes complete pseudorange and carrier phase and their discrimination thresholds, ερ and δρ represent the noise level of GNSS pseudorange and its discrimination threshold, respectively. and They represent the noise level of the GNSS carrier phase and its discrimination threshold, respectively. δT and Ts represent the data age of the PPP-RTK differential enhancement information and its discrimination threshold, respectively.

4. The method for autonomous integrity assessment of multi-source fusion positioning in complex environments according to claim 1 is characterized in that: The integrity judgment equation of the zero-value deviation in step B is: Among them, GNSS PPP-RTK / INS is the fusion solution anomaly indicator. and They represent the acceleration zero deviation stability threshold and the angular velocity zero deviation stability threshold, respectively, and b g Indicates the angular velocity zero value deviation, b a Indicates the acceleration zero value deviation, and They represent the virtual observation values ​​of the acceleration zero deviation and angular velocity zero deviation at the previous moment respectively; in, and They represent the constraint matrices of the virtual observation equations for the zero-value deviation of acceleration and the zero-value deviation of angular velocity at the previous moment, and They represent the acceleration zero-value deviation measurement error and angular velocity zero-value deviation measurement error at the previous moment respectively.

5. The method for autonomous integrity assessment of multi-source fusion positioning in complex environments according to claim 1 is characterized in that: The position and attitude combined lateral integrity assessment method in step C includes the following steps: C.1 Identify abnormal status information of each sensor position and posture: Among them, Idv i is the abnormal position and posture mark of each sensor, r med and Rs. med The position and attitude rotation matrices are obtained based on median filtering at the current moment, r i and Rs. i The position and attitude rotation matrices obtained by each sensor, i can be PPP_RTK_INS or INS or SLAM; Among them, Median is the median filter function, ε r and ε Rs They are respectively expressed as the position deviation judgment threshold and the attitude rotation matrix deviation judgment threshold; C.2 Using the above sensor position and posture abnormality identifier Idv i Based on, select the pose identifier Idv i The position and attitude matrix of each sensor with a value of 1 is used as the observation data, and the graph optimization fusion positioning method is adopted to obtain the position and attitude of the intelligent rescue equipment after GNSS PPP-RTK / INS / laser SLAM fusion.

6. The method for autonomous integrity assessment of multi-source fusion positioning in complex environments according to claim 1 is characterized in that: The zero-value deviation correctness evaluation equation in step D is: Among them, Idx bias Update the correctness flag for zero-value deviations.

7. The method for autonomous integrity assessment of multi-source fusion positioning in complex environments according to claim 6 is characterized in that: In step D, updating the correctness flag according to the zero-value deviation, and updating the inertial navigation acceleration zero-value deviation and the angular velocity zero-value deviation include the following steps:

Images