Error correction method and system for ultra-wideband ranging

Abstract

The invention discloses an ultra-wideband ranging error correction method and system, and relates to the technical field of ultra-wideband ranging. The problem that an intelligent driving system mistakenly evaluates the distance and the position of a surrounding obstacle due to the fact that measurement errors exist when a bidirectional distance measurement method is adopted in the ultra-wideband distance measurement technology is solved. The method comprises the following steps: S1, analyzing error sources according to the principle of an ultra-wideband signal bidirectional ranging method, wherein the error sources comprise clock drift, hardware delay and signal power; s2, correcting clock drift according to a linear interpolation method; s3, constructing an error compensation model; and S4, correcting errors caused by hardware delay and signal power according to the error compensation model. The method is applied to the field of intelligent driving.

Description

Technical Field

[0001] The present invention relates to the technical field of ultra-wideband ranging, and in particular to an error correction method for ultra-wideband ranging. Background Art

[0002] In recent years, cars have become an important part of people's lives. They have entered thousands of households and brought convenience to people's lives. However, with the increase in the number of cars, the incidence of traffic accidents has also increased. Drivers alone cannot accurately perceive road traffic conditions and avoid danger in an emergency. Once a traffic accident occurs, it will cause casualties and property losses. The application of intelligent driving technology can effectively reduce the incidence of traffic accidents and ensure road safety.

[0003] At present, intelligent driving systems often use millimeter-wave radar or laser radar to complete the ranging of surrounding targets. However, millimeter-wave radar has low resolution and weak detection capabilities for small obstacles, animals and other small targets. In addition, its detection angle is limited, and there are blind spots when detecting the side of the vehicle and other directions. Laser radar is expensive and sensitive to bad weather such as rain and fog, which affects the detection effect of targets around the car. Ultra-wideband ranging technology does not rely on carrier signals for propagation, and has the characteristics of wide bandwidth and short pulses. It has high time resolution, strong anti-interference ability, and decimeter-level positioning accuracy. Therefore, it can provide stable positioning and ranging services in complex intelligent driving environments. In addition, ultra-wideband has low-latency data transmission capabilities, can transmit and process environmental data in real time, enable intelligent driving systems to respond quickly, and enhance the safety and reliability of intelligent driving.

[0004] Ultra-wideband ranging technology often uses the arrival time method to calculate the distance, that is, under the premise of knowing the signal propagation speed, the time it takes for the signal to be sent and received is recorded, so as to accurately calculate the distance between the sender and the receiver. In order to ensure that the sender and the receiver have the same time reference, calculating the distance based on the arrival time requires the equipment to run at the same frequency, but due to the limitation of processing accuracy, the frequency of the crystal oscillator will be slightly different, affecting the accuracy of the ranging. The current method with higher accuracy is the two-way ranging method. This method does not require the equipment to run at the same frequency and has higher accuracy than the arrival time method, but this method always causes measurement errors. When the measurement error is large, the intelligent driving system will incorrectly evaluate the distance and position of the surrounding obstacles, resulting in wrong decisions; at the same time, the error in ranging will also lead to inaccurate path planning, causing the vehicle to choose an unsafe driving route or unnecessary braking. Summary of the invention

[0005] Aiming at the problem that the ultra-wideband ranging technology uses a two-way ranging method with measurement errors, which causes the intelligent driving system to incorrectly evaluate the distance and position of surrounding obstacles, the present invention proposes a method for correcting ultra-wideband ranging errors, the method comprising:

[0006] S1: Analyze the error sources according to the principle of the ultra-wideband signal two-way ranging method, the error sources include clock drift, hardware delay and signal power;

[0007] S2: Correct the clock drift according to the linear interpolation method;

[0008] S3: construct error compensation model;

[0009] S4: Correct the errors caused by hardware delay and signal power according to the error compensation model.

[0010] Furthermore, a preferred method is proposed, wherein step S1 comprises:

[0011] S11: Calculate the time difference between the reference station sending and receiving signals;

[0012] S12: Calculating the flight time in the two-way ranging method according to the time difference between the transmitted and received signals;

[0013] S13: Calculating the distance between the signal transmitting device and the signal receiving device according to the flight time;

[0014] S14: Calculate the existing clock drift error according to the optional method AltDS-TWR of bilateral distance measurement;

[0015] S15: Analyze error influencing factors according to the clock drift error. The error influencing factors include power impact and hardware delay.

[0016] Furthermore, a preferred method is proposed, wherein step S14 comprises:

[0017] Calculation of the flight time T based on the alternative method AltDS-TWR for bilateral distance measurement AltDS :

[0018]

[0019] in, It represents the time difference between the first reception and the first transmission of the reference station. Indicates the time difference between the first transmission and the first reception of the tag signal; Indicates the time difference between the second signal received by the tag and the first signal transmitted. It represents the time difference between the second transmission and the first reception of the signal by the reference station;

[0020] Time of flight with frequency offset for:

[0021]

[0022] Among them, eR is the clock drift of the reference station tag;

[0023] Clock drift of AltDS-TWR ε AltDS for:

[0024]

[0025] Furthermore, a preferred method is proposed, wherein the error compensation model of step S3 includes 1 input layer, 2 hidden layers and 1 output layer; the input layer is the first part of the neural network, used to receive data and pass it to the next layer, and the number of nodes in the input layer is based on the dimension of the input data; the hidden layer is used to extract and combine the features of the input data to generate a higher-level representation; the hidden layer includes 2 fully connected layers to process high-dimensional features, and each hidden layer contains 64 neurons; the output layer is responsible for outputting data calibrated by the neural network model, using a linear function as an activation function.

[0026] Furthermore, a preferred embodiment is proposed, in which the input layer is provided with 1 node.

[0027] Furthermore, a preferred embodiment is proposed, in which a ReLU function is used as an activation function in the hidden layer.

[0028] Furthermore, a preferred embodiment is proposed, wherein the method further comprises training an error compensation model, specifically:

[0029] The UWB signal is generated by the DW1000 transceiver and distributed to the circularly polarized patch antenna according to the power divider and transmitted. The UWB ranging value is obtained and used as the measured value, and the total station ranging value is used as the true value;

[0030] The measured values ​​and true values ​​are input into the neural network model for training.

[0031] Based on the same inventive concept, the present invention also proposes a system for ultra-wideband ranging error correction, the system comprising:

[0032] An error source analysis unit, used for analyzing error sources of ultra-wideband signals according to a two-way ranging method, wherein the error sources include clock drift, hardware delay and signal power;

[0033] A clock drift correction unit, used to correct the clock drift according to a linear interpolation method;

[0034] An error compensation model building unit, used for building an error compensation model;

[0035] The error correction unit is used to correct the error caused by hardware delay and signal power according to the error compensation model.

[0036] Based on the same inventive concept, the present invention also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for ultra-wideband ranging error correction according to any one of the above.

[0037] Based on the same inventive concept, the present invention also proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of a method for ultra-wideband ranging error correction as described in any one of the above items are executed.

[0038] The present invention is beneficial in that:

[0039] The present invention proposes a method for correcting errors in ultra-wideband ranging. The method first analyzes various sources of errors that affect ranging accuracy based on the principle of the distance calculation method. Afterwards, a linear interpolation correction method is used to address the clock drift problem in the error. The method can better adapt to the linear change trend of clock drift. This step effectively reduces the ranging error caused by clock instability by compensating for the slight differences between time references. Then, an error compensation model is constructed to calibrate the errors caused by hardware delays and signal power changes. The model can automatically learn from a large amount of actual data and identify the nonlinear effects of hardware delays and signal power changes on ranging results. By continuously optimizing model parameters and structures, accurate prediction and compensation of these two error factors are successfully achieved. Finally, by combining linear interpolation correction with neural network model optimization, not only the accuracy of ultra-wideband ranging is significantly improved, but also its adaptability and stability in complex and changing environments are enhanced.

[0040] The present invention proposes a method for correcting errors in ultra-wideband ranging, which systematically considers multiple sources of errors, can greatly improve the ranging accuracy, and ensure that the intelligent driving system can more accurately identify surrounding obstacles. By establishing an error compensation model, the method can be adaptively adjusted to different working environments and conditions, thereby improving the robustness of the system in a variety of scenarios. Using a linear interpolation method to correct clock drift is a relatively simple and effective method that can be easily integrated into existing ultra-wideband ranging systems without the need for large-scale hardware modifications. Accurate distance and position assessment is crucial to the safety of autonomous driving and assisted driving systems, and reducing measurement errors helps improve the vehicle's responsiveness and driving safety. Through the optimization and correction of software methods, the reliance on hardware costs can be reduced, thereby reducing the overall manufacturing and maintenance costs of the intelligent driving system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flow chart of a method for ultra-wideband ranging error correction according to implementation mode 1;

[0042] Figure 2 This is a schematic diagram of the two-way ranging theory described in Implementation Mode 11;

[0043] Figure 3 This is a schematic diagram of the structure of the error compensation model according to the eleventh embodiment;

[0044] Figure 4 This is a schematic diagram of the ReLU function described in Implementation Example 11;

[0045] Figure 5 This is a flowchart of the error compensation model training described in Implementation Method 11. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0047] Implementation method 1, see Figure 1 The present embodiment describes a method for correcting an error in ultra-wideband ranging, the method comprising:

[0048] S1: Analyze the error sources according to the principle of the ultra-wideband signal two-way ranging method, the error sources include clock drift, hardware delay and signal power;

[0049] S2: Correct the clock drift according to the linear interpolation method;

[0050] S3: construct error compensation model;

[0051] S4: Correct the errors caused by hardware delay and signal power according to the error compensation model.

[0052] In the ranging process, the transmission and reception of signals rely on accurate clocks. If the clocks between the transmitter and the receiver are not synchronized, then even if the signal propagation time is correct, the actual measured distance will still have errors. Hardware components (such as transmitters and receivers) will introduce certain delays in the process of signal transmission and reception, which will affect the measurement of distance. Signals will attenuate during propagation, resulting in a weakening of the received signal strength, which will affect the accuracy of the ranging algorithm. Using the linear interpolation method, the time drift value is estimated through the linear relationship between two known data points, and the time drift generated in the ranging process is corrected to ensure that the ranging time is closer to the actual value, reduce measurement errors, prevent error accumulation, and improve ranging accuracy. By analyzing the source of measurement errors, an error compensation model is constructed, which can compensate for errors caused by hardware delays and signal power changes. The incoming UWB data is processed by the error compensation model to eliminate the ranging errors caused by hardware delays and signal power in the data, thereby improving ranging accuracy.

[0053] The correction method proposed in this embodiment systematically considers multiple sources of errors, which can greatly improve the ranging accuracy and ensure that the intelligent driving system can more accurately identify surrounding obstacles. By establishing an error compensation model, the method can be adaptively adjusted to different working environments and conditions, thereby improving the robustness of the system in various scenarios. Using linear interpolation method for clock drift correction is a relatively simple and effective method that can be easily integrated into existing ultra-wideband ranging systems without large-scale hardware modifications. Accurate distance and position assessment is crucial to the safety of autonomous driving and assisted driving systems. Reducing measurement errors helps improve the vehicle's responsiveness and driving safety. Through the optimization and correction of software methods, the reliance on hardware costs can be reduced, thereby reducing the overall manufacturing and maintenance costs of the intelligent driving system.

[0054] Implementation method 2: This implementation method further limits the method for ultra-wideband ranging error correction described in implementation method 1, and step S1 includes:

[0055] S11: Calculate the time difference between the reference station sending and receiving signals;

[0056] S12: Calculating the flight time in the two-way ranging method according to the time difference between the transmitted and received signals;

[0057] S13: Calculating the distance between the signal transmitting device and the signal receiving device according to the flight time;

[0058] S14: Calculate the existing clock drift error according to the optional method AltDS-TWR of bilateral distance measurement;

[0059] S15: Analyze error influencing factors according to the clock drift error. The error influencing factors include power impact and hardware delay.

[0060] In this embodiment, by accurately calculating the time difference between the transmission and reception of the signal, the accuracy of the ranging can be effectively improved. This process can ensure that more accurate flight time data is obtained in the two-way ranging, laying the foundation for the subsequent distance calculation. Through the optional method AltDS-TWR for bilateral distance measurement, this method has good anti-clock drift capability and can systematically determine whether there is a clock drift error. This method provides a standardized operating procedure that helps to identify and classify different types of errors at an early stage. By comprehensively analyzing multiple factors such as signal power influence and hardware delay, this process can help the system more comprehensively understand the complex error situations that may be encountered in actual applications, and provide data support for the effectiveness of subsequent correction measures.

[0061] Implementation method 3: This implementation method is a further limitation of the method for ultra-wideband ranging error correction described in implementation method 2, and step S14 includes:

[0062] Calculation of the flight time T based on the alternative method AltDS-TWR for bilateral distance measurement AltDS :

[0063]

[0064] in, Indicates the time difference between the first reception and the first transmission of the reference station. The time difference between the reception and transmission of the reference station. Indicates the time difference between the first transmission and the first reception of the tag. The time difference between the transmission and reception of the signal; Indicates the time difference between the second signal received by the tag and the first signal transmitted. It represents the time difference between the second transmission and the first reception of the signal by the reference station;

[0065] Time of flight with frequency offset for:

[0066]

[0067] Among them, e R is the clock drift of the reference station tag;

[0068] Clock drift of AltDS-TWR ε AltDS for:

[0069]

[0070] Implementation method 4. This implementation method is a further limitation of the method for ultra-wideband ranging error correction described in implementation method 1. The error compensation model of step S3 includes 1 input layer, 2 hidden layers and 1 output layer; the input layer is the first part of the neural network, which is used to receive data and pass it to the next layer, and the number of nodes in the input layer is based on the dimension of the input data; the hidden layer is used to extract and combine the features of the input data to generate a higher-level representation; the hidden layer includes 2 fully connected layers to process high-dimensional features, and each hidden layer contains 64 neurons; the output layer is responsible for outputting data calibrated by the neural network model, and uses a linear function as an activation function.

[0071] The error compensation model described in this embodiment adopts a neural network structure, especially a design including two hidden layers, which can effectively extract and learn high-dimensional features in the input data. Through the deep network architecture, the model can identify complex nonlinear relationships and improve the accuracy of ranging. The fully connected layer and nonlinear activation function of the neural network (although the output layer uses a linear function, the hidden layer is a nonlinear function) make the model more flexible and adaptable when dealing with complex error patterns. This nonlinear modeling capability helps to better fit the errors in the actual ranging data.

[0072] In practical applications, the flexible design of the number of input layer nodes makes the method scalable. The input layer nodes can be adjusted according to the newly collected feature data (such as changes in the ranging environment) without reconstructing the entire model, which improves the convenience of later applications. The final output layer is calibrated by the neural network model, and the error compensation obtained based on model learning is used to improve the accuracy of the ranging data and ensure that the application results meet actual needs.

[0073] Implementation mode 5: This implementation mode is a further limitation of the method for ultra-wideband ranging error correction described in implementation mode 4, and the input layer is provided with one node.

[0074] Implementation method 6: This implementation method is a further limitation of the method for ultra-wideband ranging error correction described in implementation method 4, and the ReLU function is used as the activation function in the hidden layer.

[0075] Implementation 7: This implementation is a further limitation of the method for ultra-wideband ranging error correction described in Implementation 1. The method also includes a training error compensation model, specifically:

[0076] The UWB signal is generated by the DW1000 transceiver and distributed to the circularly polarized patch antenna according to the power divider and transmitted. The UWB ranging value is obtained and used as the measured value, and the total station ranging value is used as the true value;

[0077] The measured values ​​and true values ​​are input into the neural network model for training.

[0078] Embodiment 8: A system for ultra-wideband ranging error correction described in this embodiment includes:

[0079] An error source analysis unit, used for analyzing error sources of ultra-wideband signals according to a two-way ranging method, wherein the error sources include clock drift, hardware delay and signal power;

[0080] A clock drift correction unit, used to correct the clock drift according to a linear interpolation method;

[0081] An error compensation model building unit, used for building an error compensation model;

[0082] The error correction unit is used to correct the error caused by hardware delay and signal power according to the error compensation model.

[0083] Embodiment 9. A computer device described in this embodiment includes a memory and a processor, wherein a computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes a method for ultra-wideband ranging error correction according to any one of embodiments 1 to 7.

[0084] Embodiment 10: A computer-readable storage medium described in this embodiment, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a method for ultra-wideband ranging error correction as described in any one of embodiments 1 to 7 are executed.

[0085] Implementation Method 11: See Figures 2 to 5 This embodiment provides a specific example of the method for ultra-wideband ranging error correction described in the first embodiment, and is also used to explain the second to seventh embodiments, specifically:

[0086] Firstly, the error sources are analyzed according to the distance calculation method. It is found that the two-way ranging method will cause errors due to clock drift, hardware delay and signal power. After correcting the clock drift according to the linear interpolation method, an error compensation model is constructed to correct the errors caused by hardware delay and signal power. In this way, the accuracy of ultra-wideband ranging can be improved.

[0087] Specifically, the first step in the present invention is to analyze the source of error according to the ranging method. In the present invention, the ultra-wideband (UWB) signal measures the target distance through a two-way ranging method, which does not require the signal transmitting device and the signal receiving device to work at the same frequency, is not affected by the time synchronization between nodes, and can provide more accurate distance information.

[0088] The distance measurement method between UWB signal transceiver devices is as follows: Figure 2 In this embodiment, the time when the reference station transmits or receives the signal is set to T R , set the time when the tag transmits or receives the signal to T T , TOF represents the flight time of the signal. The reference station is at T1 R The tag transmits a signal at T1 T The tag receives the signal at T2 T The reference station transmits a signal at T2 R The reference station receives the signal at T3 R The tag emits a signal at T3 T Receive signal when.

[0089] The formula for calculating the time difference between sending and receiving signals is:

[0090]

[0091] in, It represents the time difference between the reference station receiving and transmitting signals. Indicates the time difference between the tag transmitting and receiving signals.

[0092] Therefore, the calculation formula of Time of Flight (TOF) in the two-way ranging method is:

[0093]

[0094] The distance between the signal transmitting and signal receiving devices is calculated as:

[0095] d=TOF×c (4)

[0096] Among them, d represents the distance between the signal transmitting and signal receiving devices, and c represents the speed of light.

[0097] In a single measurement, the clock drift between the reference station and the tag is defined as e R and e T It is assumed that the clock drift is constant during a single measurement. In order to limit the impact of clock drift, this method applies an alternative method of bilateral distance measurement (Alternative Double-Sided Two-Way Ranging, AltDS-TWR), which has better resistance to clock drift.

[0098] The AltDS-TWR method exploits the fact that the time difference between the first and last message at both the transmitter and the receiver must be the same, i.e. The difference between the two is the clock drift error, which can be corrected by linear interpolation. The TOF calculation formula of the AltDS-TWR method is:

[0099]

[0100] The TOF calculation formula with frequency offset is:

[0101]

[0102] The clock drift of AltDS-TWR is given by the formula The calculation results are:

[0103] ε AltDS =e R T AltDS (7)

[0104] The timestamp of the signal is affected by the signal power. An increase in signal power leads to a negative shift in the timestamp. Conversely, a decrease in signal power leads to a positive shift in the timestamp. Signal power attenuation also has a nonlinear effect on the measurement. In addition, the signal reaches the antenna and is transmitted through the power divider and transmission line. The hardware delay will add a fixed error Δ to the TOF. delay , which also causes a delay in setting the timestamp.

[0105] Next, in this implementation, an error compensation model is constructed to calibrate the ultra-wideband ranging hardware delay and signal power error to improve the ranging accuracy.

[0106] By modeling the distance measurement error, the distance measurement value can be effectively compensated, thereby improving the accuracy of ultra-wideband ranging positioning. Mathematical models commonly used for fitting functions include polynomial function models, exponential function models, logarithmic function models, and trigonometric function models. However, these models are traditional mathematical models and cannot adapt to changing environments. In actual situations, the distribution of UWB distance measurement errors has irregular characteristics, and traditional fitting function models cannot effectively eliminate the impact of distance measurement errors.

[0107] Due to the rapid development of deep learning technology, the neural network model can be trained using sampled data through its powerful nonlinear fitting ability, and a more accurate ranging error correction model can be constructed to improve positioning accuracy. For signal power and hardware delay errors, this method constructs an error compensation model based on a neural network.

[0108] Neural network models such as Figure 3 As shown in Figure 3, the model has 1 input layer, 2 hidden layers, and 1 output layer.

[0109] The input layer is the first part of the neural network, which is responsible for receiving data and passing it to the next layer. The number of nodes in the input layer is based on the dimension of the input data. In this method, the input UWB ranging value is one-dimensional data, so the input layer is set with 1 node.

[0110] The hidden layer is responsible for extracting and combining the features of the input data to generate a higher-level representation. To ensure global combination of features, this method designs two fully connected layers in the hidden layer to process high-dimensional features. Each hidden layer contains 64 neurons, which balances the model's expressive power and computational complexity. The design of the two hidden layers is deep enough to capture complex feature relationships, but not too complex to be difficult to train. This design can gradually extract and combine features.

[0111] In the hidden layer, the ReLU (Rectified Linear Unit) function is used as the activation function, and its expression is:

[0112] f(x)=max(0,x) (8)

[0113] The schematic diagram of the ReLU function is as follows Figure 4 As shown in the figure, the activation function introduces nonlinearity, enhances the expressiveness of the model, and alleviates the gradient vanishing problem, enabling stable training of the neural network.

[0114] The output layer is responsible for outputting the data calibrated by the neural network model. This layer uses a linear function as the activation function and outputs the calibrated one-dimensional UWB ranging value through a node.

[0115] The UWB signal is generated by DW1000 and distributed to the circularly polarized patch antenna through a power divider and transmitted. The UWB ranging value is used as the measured value, the total station ranging value is used as the true value, and the measured value and the true value are input into the error compensation model for training. The error compensation model training process designed in this embodiment is as follows: Figure 5 As shown, specifically including:

[0116] Preprocess the UWB ranging value, such as normalization and denoising;

[0117] Initialize the weights and biases in the neural network;

[0118] Using the training dataset to train the error compensation model;

[0119] Calculate the output of each node in hidden layer 1; Calculate the output of each node in hidden layer 2;

[0120] The data processed by the hidden layer will be passed to the output layer, and the output layer will give the final prediction result;

[0121] The gap between the model's prediction results and the true value is evaluated by calculating the loss function;

[0122] Determine whether the loss function converges or the number of training times reaches the maximum value; if the value of the loss function no longer decreases significantly, or the number of training times reaches the preset maximum value, the training process ends;

[0123] The gradient is calculated by the loss function, which indicates the direction in which the loss function decreases fastest in the parameter space;

[0124] Update weights and biases through the gradient descent algorithm to reduce the value of the loss function;

[0125] Continue training the sample data using the weights and biases of the neural network obtained in the previous step until the loss function converges or the number of training times reaches the maximum value;

[0126] After training, the UWB ranging value is input into the network, and the output is the calibrated data. The model can compensate for nonlinear and linear errors.

[0127] The advantages of this embodiment are as follows:

[0128] (1) Using the AltDS-TWR method to measure distance using UWB signals: The AltDS-TWR method does not require devices to work at the same frequency, which reduces the requirements for device crystal oscillator accuracy, reduces error sources, and improves the accuracy of UWB signal distance measurement.

[0129] (2) Use linear interpolation method to correct time drift: Use linear interpolation method to estimate the time drift value through the linear relationship between two known data points, and correct the time drift generated during the ranging process to ensure that the ranging time is closer to the actual value, reduce measurement errors, prevent error accumulation, and improve ranging accuracy.

[0130] (3) Constructing and training the error compensation model: By designing the structure of the input layer, hidden layer, and output layer and their connection relationship, a neural network model consisting of 1 input layer, 2 hidden layers, and 1 output layer is constructed. The error compensation model is trained to make the neural network meet the application requirements.

[0131] (4) Calibration of hardware delay and signal power during ranging based on the error compensation model: The incoming UWB data is processed through deep learning methods to eliminate the ranging errors caused by hardware delay and signal power in the data, thereby improving the ranging accuracy.

[0132] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit its protection scope. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that after reading the present invention, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the pending claims of the present invention.

Claims

1. A method for correcting ultra-wideband ranging errors, characterized in that: The method comprises: S1: Analyze the error sources according to the principle of the ultra-wideband signal two-way ranging method, the error sources include clock drift, hardware delay and signal power; S2: Correct the clock drift according to the linear interpolation method; S3: construct error compensation model; S4: Correct the errors caused by hardware delay and signal power according to the error compensation model.

2. The method for correcting ultra-wideband ranging errors according to claim 1, characterized in that: The step S1 comprises: S11: Calculate the time difference between the reference station sending and receiving signals; S12: Calculating the flight time in the two-way ranging method according to the time difference between the transmitted and received signals; S13: Calculating the distance between the signal transmitting device and the signal receiving device according to the flight time; S14: Calculate the existing clock drift error according to the optional method AltDS-TWR of bilateral distance measurement; S15: Analyze error influencing factors according to the clock drift error. The error influencing factors include power impact and hardware delay.

3. The method for correcting ultra-wideband ranging errors according to claim 2, characterized in that: The step S14 comprises: Calculation of the flight time T based on the alternative method AltDS-TWR for bilateral distance measurement AltDS : in, It represents the time difference between the first reception and the first transmission of the reference station. Indicates the time difference between the first transmission and the first reception of the tag signal; Indicates the time difference between the second signal received by the tag and the first signal transmitted. It represents the time difference between the second transmission and the first reception of the signal by the reference station; Time of flight with frequency offset for: Among them, e R is the clock drift of the reference station tag; Clock drift of AltDS-TWR∈ AltDS for:

4. The method for correcting ultra-wideband ranging errors according to claim 1, characterized in that: The error compensation model of step S3 includes 1 input layer, 2 hidden layers and 1 output layer; the input layer is the first part of the neural network, used to receive data and pass it to the next layer, and the number of nodes in the input layer is based on the dimension of the input data; the hidden layer is used to extract and combine the features of the input data to generate a higher-level representation; the hidden layer includes 2 fully connected layers to process high-dimensional features, and each hidden layer contains 64 neurons; the output layer is responsible for outputting data calibrated by the neural network model, using a linear function as an activation function.

5. The method for correcting ultra-wideband ranging errors according to claim 4, characterized in that: The input layer is provided with 1 node.

6. The method for correcting ultra-wideband ranging errors according to claim 4, characterized in that: The ReLU function is used as the activation function in the hidden layer.

7. The method for correcting ultra-wideband ranging errors according to claim 1, characterized in that: The method further includes training an error compensation model, specifically: The UWB signal is generated by the DW1000 transceiver and distributed to the circularly polarized patch antenna according to the power divider and transmitted. The UWB ranging value is obtained and used as the measured value, and the total station ranging value is used as the true value; The measured values ​​and true values ​​are input into the neural network model for training.

8. A system for ultra-wideband ranging error correction, characterized in that: The system comprises: An error source analysis unit, used for analyzing error sources of ultra-wideband signals according to a two-way ranging method, wherein the error sources include clock drift, hardware delay and signal power; A clock drift correction unit, used to correct the clock drift according to a linear interpolation method; An error compensation model building unit, used for building an error compensation model; The error correction unit is used to correct the error caused by hardware delay and signal power according to the error compensation model.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes a method for correcting ultra-wideband ranging errors according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a method for correcting ultra-wideband ranging errors as claimed in any one of claims 1 to 7 are executed.

Images