Adaptive Vehicle Switching State Estimation Method and System

Abstract

The present invention provides a method and system for estimating the switching state of an adaptive vehicle, including: collecting vehicle acceleration and wheel speed; obtaining a first road slope by using the vehicle acceleration and wheel speed through a kinematic equation; obtaining GPS speed through a GPS data receiver and calculating a second road slope according to the GPS speed; updating a prediction process model through the first road slope and the second road slope to obtain the road slope at the current moment; compensating the vertical speed on the slope road surface through the road slope at the current moment; obtaining the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model; determining vehicle speed, yaw rate, and sideslip angle according to the kinematic model or the dynamic model; designing a prediction function of an observer according to the vehicle speed, yaw rate, and sideslip angle; and correcting the prediction function of the observer through the vehicle position information obtained by a GPS module to obtain vehicle longitudinal and lateral position information and vehicle body state information.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicle design and manufacturing, and more particularly to a method and system for estimating the switching state of an adaptive vehicle. Background Art

[0002] With the continuous development of automotive technology, the intelligence and automation of modern vehicles have been significantly improved. To ensure the safety and stability of vehicles under various driving conditions, accurate vehicle state estimation has become one of the key technologies. Vehicle state estimation usually involves the real-time monitoring and analysis of parameters such as vehicle position, speed, acceleration, and attitude, and this information is crucial for vehicle navigation, control, and fault detection applications.

[0003] Currently, vehicle state estimation mainly relies on the Global Positioning System (GPS) and in-vehicle sensors such as accelerometers and gyroscopes. However, the data of a single GPS or in-vehicle sensor is often affected by environmental factors, resulting in insufficient estimation accuracy. For this reason, the method of vehicle state estimation by fusing multi-source information has become a research hotspot.

[0004] Traditional vehicle state estimation methods face many limitations. First, as an important tool for vehicle positioning, GPS provides position information with relatively high accuracy. However, GPS signals are vulnerable to occlusion and interference. For example, in an environment with high-rise buildings, tunnels, underground parking lots, etc. in the city, the GPS signal may become weak or completely lost. In this case, relying solely on GPS for vehicle state estimation will become unreliable.

[0005] Secondly, in-vehicle sensors such as accelerometers and gyroscopes can provide information on vehicle acceleration and angular velocity. However, the data of these sensors are often affected by noise and drift. Especially after long-term use, the errors will gradually accumulate, resulting in inaccurate state estimation. In addition, factors such as the installation position of the sensors and the vibration of the vehicle will also affect the accuracy of the data.

[0006] Finally, a single information source (such as relying solely on GPS or in-vehicle sensors) often fails to meet the high-precision state estimation requirements in complex driving environments. For example, on highways, the speed and direction of vehicles change less, and the data of GPS and in-vehicle sensors are relatively reliable; but in complex urban environments, vehicles frequently turn, brake, and accelerate, and a single information source may not accurately reflect the actual state of the vehicle.

[0007] To overcome the limitations of a single information source, the data from different sensors are comprehensively processed to improve the accuracy and reliability of state estimation. Common methods include Kalman filtering, extended Kalman filtering, particle filtering, and adaptive filtering methods that fuse neural network and machine learning technologies.

[0008] Kalman filtering is an optimal estimation method based on linear systems and is applicable to Gaussian noise environments. For non-linear systems, the extended Kalman filter approximately achieves optimal estimation by linearizing the non-linear system.

[0009] The Kalman filtering method can effectively fuse GPS and vehicle sensor data to provide a relatively accurate vehicle state estimation. However, the performance of these methods will decline when dealing with strong non-linear and non-Gaussian noise environments.

[0010] Particle filtering is a Bayesian filtering method based on Monte Carlo simulation and is applicable to non-linear and non-Gaussian noise environments. Particle filtering approximates the posterior distribution of the system by generating a large number of samples (particles) and weighting the particles according to the observed data. Particle filtering shows good robustness when dealing with complex environments and high-dynamic systems, but has a large computational load and relatively poor real-time performance.

[0011] With the improvement of computing power and the development of machine learning technologies, adaptive filtering and neural network-based estimation methods have gradually received attention. Adaptive filtering can improve the estimation accuracy by dynamically adjusting the filtering parameters to adapt to different environments and driving conditions. Neural networks and deep learning technologies can learn the complex relationships between different sensor data through the training of a large amount of data, thereby achieving high-precision state estimation. However, these methods rely on a large amount of training data and computing resources, and the accuracy and real-time performance need to be weighed in practical applications. Summary of the Invention

[0012] In view of this, the purpose of the present invention is to provide a method and system for estimating the switching state of an adaptive vehicle, which can improve the accuracy and reliability of vehicle state estimation, effectively avoid the uncertainty brought by a single information source; ensure high-efficiency and accurate state estimation in various complex scenarios, and improve the robustness and stability of the system; improve the real-time performance of state estimation.

[0013] In the first aspect, an embodiment of the present invention provides a method for estimating the switching state of an adaptive vehicle, the method comprising:

[0014] Collecting vehicle acceleration through a longitudinal acceleration sensor and collecting wheel speed through a wheel speed sensor;

[0015] Obtaining a first road slope by using the vehicle acceleration and the wheel speed through a kinematic equation;

[0016] Obtaining GPS speed through a GPS data receiver and calculating a second road slope according to the GPS speed;

[0017] Constructing a prediction process model;

[0018] Update the prediction process model based on the first road slope and the second road slope to obtain the road slope at the current moment;

[0019] Compensate the vertical speed on the slope road surface based on the road slope at the current moment;

[0020] Obtain the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model;

[0021] Determine the vehicle speed, yaw rate, and sideslip angle according to the kinematic model or the dynamic model;

[0022] Design the prediction function of the observer according to the vehicle speed, the yaw rate, and the sideslip angle;

[0023] Calibrate the prediction function of the observer through the vehicle position information obtained by the GPS module to obtain the vehicle longitudinal and lateral position information and the vehicle body state information;

[0024] Wherein, the vehicle body state information includes the calibrated vehicle speed, the calibrated yaw rate, and the calibrated sideslip angle.

[0025] Further, the method further includes:

[0026] Update the probability of the j-th candidate model through the likelihood function of the j-th candidate model to obtain the updated model probability;

[0027] Perform a weighted sum of the estimation results of the filters of all interactive multiple models according to the updated model probability to calculate the final state estimation result;

[0028] Perform a weighted sum of the estimation results of the filters of all the interactive multiple models according to the updated model probability and the final state estimation result to calculate the final covariance estimation result.

[0029] Further, the kinematic equation is implemented in the following manner:

[0030]

[0031] Where a accel is the vehicle acceleration, is the time derivative of the speed, representing the speed change rate in the XYZ directions, g is the gravitational acceleration, and sinθ is the sine value of the first road slope.

[0032] Further, obtaining the first road slope by passing the vehicle acceleration and the wheel speed through the kinematic equation includes:

[0033] Calculate the first road slope according to the following formula:

[0034]

[0035] Among them, θ is the first road slope, and V whl is the wheel speed, which is measured by the wheel speed collected by the wheel speed sensor. XYZ

[0036] Furthermore, calculating the second road slope according to the GPS speed includes:

[0037] Calculating the second road slope according to the following formula:

[0038]

[0039] Among them, θ′ is the second road slope, and tan -1 (·) is the arctangent function, is the GPS speed component in the Z direction, is the GPS speed component in the X direction, is the GPS speed component in the Y direction.

[0040] Furthermore, constructing a prediction process model includes:

[0041]

[0042] Among them, is the predicted state vector at time k based on time k - 1, is the estimated value of the acceleration in the XYZ directions, is the estimated value of the rate of change of speed in the XYZ directions, T is the discrete time, is the estimated value of the second road slope, and sin -1 (·) is the non - sine function, and cos(·) is the cosine function.

[0043] Furthermore, the discrete process model of the kinematic model is realized by the following method:

[0044]

[0045] Among them, l f is the distance from the front axle to the vehicle center of gravity, and l r is the distance from the rear axle to the vehicle center of gravity, δ SAS is the tire corner, is the estimated value of the rate of change of speed in the XYZ directions, is the estimated value of the vehicle speed V XY is the estimated value of the yaw rate ψ, is the estimated value of the sideslip angle β, is the estimated value of yaw rate, and T is the discrete time.

[0046] Further, the discrete process model of the dynamic model is implemented in the following manner:

[0047]

[0048] where m is the vehicle mass, g is the gravitational potential energy, and I z is the yaw moment of inertia, C f is the cornering stiffness of the front tire, and C r is the cornering stiffness of the rear tire.

[0049] Further, the prediction function of the observer is implemented in the following manner:

[0050]

[0051] P k∣k-1 = F k-1 P k-1∣k-1 (F k-1 ) T + Q k-1

[0052]

[0053] where f(·) is the system nonlinear equation, T is the discrete time, and are the estimated values of the vehicle position information, is the estimated value of the vehicle speed V XY and is the estimated value of the yaw rate ψ, is the estimated value of the sideslip angle β, and P k∣k-1 , P k-1∣k-1 are the covariance, and Q k-1 is the system process noise.

[0054] [[ID=6,3]]In a second aspect, an embodiment of the present invention provides a switching state estimation system for an adaptive vehicle, and the system includes:

[0055] An acquisition module, configured to acquire vehicle acceleration through a longitudinal acceleration sensor and acquire wheel speed through a wheel speed sensor;

[0056] A first road slope calculation module, configured to obtain a first road slope by using the vehicle acceleration and the wheel speed through a kinematic equation;

[0057] A second road slope calculation module, configured to obtain GPS speed through a GPS data receiver and calculate a second road slope according to the GPS speed;

[0058] A building module for building a prediction process model;

[0059] An update module for updating the prediction process model by using the first road slope and the second road slope to obtain the road slope at the current moment;

[0060] A compensation module for compensating the vertical speed on the slope road surface by using the road slope at the current moment;

[0061] An acquisition module for acquiring the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model;

[0062] A determination module for determining the vehicle speed, yaw rate and sideslip angle according to the kinematic model or the dynamic model;

[0063] A design module for designing the prediction function of the observer according to the vehicle speed, the yaw rate and the sideslip angle;

[0064] A calibration module for calibrating the prediction function of the observer by using the vehicle position information obtained by the GPS module to obtain the vehicle longitudinal and lateral position information and the vehicle body state information;

[0065] Wherein, the vehicle body state information includes the calibrated vehicle speed, the calibrated yaw rate and the calibrated sideslip angle.

[0066] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, where a computer program that can run on the processor is stored on the memory, and when the processor executes the computer program, the above-mentioned adaptive vehicle switching state estimation method is implemented.

[0067] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having non-volatile program code executable by a processor, and the program code causes the processor to execute the above-mentioned adaptive vehicle switching state estimation method.

[0068] Embodiments of the present invention provide a method and system for estimating the switching state of an adaptive vehicle, including: collecting vehicle acceleration through a longitudinal acceleration sensor and collecting wheel speed through a wheel speed sensor; obtaining a first road slope by using the vehicle acceleration and the wheel speed through a kinematic equation; obtaining GPS speed through a GPS data receiver and calculating a second road slope according to the GPS speed; constructing a prediction process model; updating the prediction process model by using the first road slope and the second road slope to obtain the road slope at the current moment; compensating the vertical speed on the sloped road surface by using the road slope at the current moment; obtaining the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model; determining vehicle speed, yaw rate, and sideslip angle according to the kinematic model or the dynamic model; designing a prediction function of an observer according to the vehicle speed, yaw rate, and sideslip angle; correcting the prediction function of the observer by using the vehicle position information obtained by a GPS module to obtain vehicle transverse and longitudinal position information and vehicle body state information; where the vehicle body state information includes the corrected vehicle speed, the corrected yaw rate, and the corrected sideslip angle; it can improve the accuracy and reliability of vehicle state estimation, effectively avoid the uncertainty brought by a single information source; ensure efficient and accurate state estimation still in various complex scenarios, and enhance the robustness and stability of the system; improve the real-time performance of state estimation.

[0069] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, the claims, and the drawings.

[0070] To make the above objectives, features, and advantages of the present invention more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0072] Figure 1 It is a flowchart of a method for estimating the switching state of an adaptive vehicle provided in Embodiment 1 of the present invention;

[0073] Figure 2 It is a schematic diagram of position estimation in a two-dimensional plane provided in Embodiment 1 of the present invention;

[0074] Figure 3Schematic diagram of the switching state estimation process of the adaptive vehicle integrating GPS and vehicle sensor information provided by Embodiment 1 of the present invention;

[0075] Figure 4 Schematic diagram of the switching state estimation system of the adaptive vehicle provided by Embodiment 2 of the present invention;

[0076] Figure 5 Schematic diagram of the structure of the electronic device provided by Embodiment 3 of the present invention. Detailed implementation manners

[0077] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0078] The present application designs a longitudinal vehicle state estimation module for estimating the longitudinal speed and road slope of the vehicle. The road slope estimation is used to compensate the vertical speed on the slope road surface. A lateral vehicle state estimation module is designed based on an interactive multiple model filter for estimating the yaw rate, yaw angle and sideslip angle of the vehicle. A position estimation module is also designed to fuse the estimation results of the longitudinal vehicle state estimation module and the lateral vehicle state estimation module with the GPS data to obtain more accurate position information.

[0079] Specifically: First, the vehicle acceleration is collected through a longitudinal acceleration sensor, and the wheel speed is collected through a wheel speed sensor; the vehicle acceleration and the wheel speed are used in the kinematic equation to obtain the first road slope; the GPS speed is obtained through a GPS data receiver, and the second road slope is calculated according to the GPS speed; a prediction process model is constructed; the prediction process model is updated by the first road slope and the second road slope to obtain the road slope at the current moment; the vertical speed on the slope road surface is compensated by the road slope at the current moment.

[0080] Then, the j-th candidate model is obtained. The j-th candidate model is a kinematic model or a dynamic model; the vehicle speed, yaw rate and sideslip angle are determined according to the kinematic model or the dynamic model; the prediction function of the observer is designed according to the vehicle speed, yaw rate and sideslip angle; the prediction function of the observer is corrected by the vehicle position information obtained by the GPS module to obtain the vehicle longitudinal and lateral position information and the vehicle body state information; wherein, the vehicle body state information includes the corrected vehicle speed, the corrected yaw rate and the corrected sideslip angle.

[0081] The adaptive vehicle state estimation method that fuses GPS and in-vehicle sensor information shows great potential in practical applications. The future development trends mainly include the following aspects. First of all, in addition to traditional GPS and in-vehicle sensors, future vehicle state estimation will make more use of other sensors, such as lidar, millimeter-wave radar, cameras, etc. These sensors can provide richer environmental information, and when fused with existing sensor data, can significantly improve the accuracy and reliability of state estimation.

[0082] Secondly, the development of adaptive filtering and intelligent algorithms will make vehicle state estimation more flexible and efficient. By adjusting filtering parameters in real time and using machine learning techniques, adaptive algorithms can better handle complex and dynamic driving environments and achieve more accurate state estimation.

[0083] Thirdly, with the development of 5G and edge computing technologies, vehicle state estimation will be able to process large amounts of data more in real time. Edge computing can transfer some computing tasks from the cloud to the vehicle local, reducing data transmission latency and improving the real-time performance and response speed of the system.

[0084] Finally, with the development of intelligent connected vehicles, vehicle state estimation systems will face more and more network security challenges. Future research needs to pay more attention to the security of the system, prevent data tampering and information leakage, and ensure the reliability and security of vehicle state estimation.

[0085] In summary, the adaptive vehicle state estimation method that fuses GPS and in-vehicle sensor information is of great significance in improving vehicle driving safety, navigation accuracy, and autonomous driving ability. With the continuous progress of technology and the growth of application requirements, this field will continue to develop and provide a more solid foundation for intelligent transportation and autonomous driving technologies.

[0086] For the convenience of understanding this embodiment, the embodiments of the present invention will be introduced in detail below.

[0087] Embodiment 1:

[0088] Figure 1 It is a flowchart of the switching state estimation method for the adaptive vehicle provided in Embodiment 1 of the present invention.

[0089] Refer to Figure 1 and the method includes the following steps:

[0090] Step S101, collect vehicle acceleration through a longitudinal acceleration sensor and collect wheel speed through a wheel speed sensor;

[0091] Step S102, obtain the first road slope by using the vehicle acceleration and wheel speed through the kinematic equation;

[0092] Step S103: Obtain the GPS speed through a GPS data receiver, and calculate the second road slope based on the GPS speed;

[0093] Step S104: Construct a prediction process model;

[0094] Step S105: Update the prediction process model with the first road slope and the second road slope to obtain the road slope at the current moment;

[0095] Step S106: Compensate the vertical speed on the slope road surface with the road slope at the current moment;

[0096] Step S107: Obtain the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model;

[0097] Step S108: Determine the vehicle speed, yaw rate, and sideslip angle according to the kinematic model or the dynamic model;

[0098] Step S109: Design the prediction function of the observer based on the vehicle speed, yaw rate, and sideslip angle;

[0099] Step S110: Calibrate the prediction function of the observer with the vehicle position information obtained by the GPS module to obtain the vehicle longitudinal and lateral position information and the vehicle body state information;

[0100] Among them, the vehicle body state information includes the calibrated vehicle speed, the calibrated yaw rate, and the calibrated sideslip angle.

[0101] Further, referring to Figure 2 and Figure 3 , the first road slope can be estimated using a longitudinal acceleration sensor and a wheel speed sensor (wheel speed sensor). The longitudinal acceleration sensor measures the vehicle acceleration, the steering wheel angle sensor is used to measure the steering wheel angle, and the yaw rate sensor is used to measure the yaw rate. The kinematic equation is implemented as follows, as shown in formula (1):

[0102]

[0103] Among them, a accel is the vehicle acceleration, which is the total acceleration of the vehicle in the XYZ directions, is the time derivative of the speed, representing the speed change rate in the XYZ directions, that is, a part of the acceleration, g is the gravitational acceleration, usually about 9.81 m / s 2 , and sinθ is the sine value of the first road slope.

[0104] Further, V XYZ can use the V from the wheel speed sensor whlmeasured using the data, and the vehicle acceleration is derived by numerically differentiating the same data. To reduce the noise component, all measurements are filtered through a second-order Butterworth low-pass filter. Using the above relationships, step S102 includes:

[0105] Calculating a first road slope according to formula (2):

[0106]

[0107] where θ is the first road slope, V whl is the wheel speed, and V XYZ is measured by the wheel speed collected by the wheel speed sensor.

[0108] Furthermore, the ratio of the vertical speed to the horizontal speed of the vehicle obtained from the GPS data receiver provides an unbiased road slope estimate. The following formula (3) shows the relationship between the road slope and the speed data according to the GPS, and step S103 includes:

[0109] Calculating a second road slope according to formula (3):

[0110]

[0111] where θ′ is the second road slope, and tan -1 (·) is the arctangent function, is the GPS speed component in the Z direction, is the GPS speed component in the X direction, is the GPS speed component in the Y direction.

[0112] Furthermore, the extended Kalman filter is used to estimate the longitudinal state of the vehicle. Combining the road slope estimates from in-vehicle sensors and the GPS can generate more accurate and reliable road slope information. Then, the three-dimensional vehicle speed is converted to a two-dimensional plane using the road slope estimate. A prediction process model is constructed, including:

[0113]

[0114] where, is the predicted state vector at time k based on the predicted state vector at time k - 1, is the estimated value of the acceleration in the XYZ directions, is the estimated value of the rate of change of speed in the XYZ directions, T is the discrete time, is the estimated value of the second road slope, and sin -1 (·) is the non-sine function, and cos(·) is the cosine function.

[0115] And the measurement vector of the update step refers to formula (5):

[0116]

[0117] Among them, z is the observation vector, and H is an observation matrix, which is usually used for the state estimation of the system.

[0118] The vehicle model set of the multiple model filter includes a kinematic model and a dynamic model. The kinematic model represents the vehicle motion under low speed and low slip conditions, because this model ignores the uncertainty of the tire lateral force. However, under high-speed and high-slip driving conditions with large lateral forces, due to the consideration of the tire lateral force, the dynamic model has better estimation performance.

[0119] The interacting multiple model filter is used to integrate two single model filters: the CKF of the kinematic model and the linear Kalman filter of the dynamic model. The process model of each filter is defined as follows:

[0120] The discrete process model of the kinematic model is implemented as follows, as shown in Equation (6):

[0121]

[0122] Among them, l f is the distance from the front axle to the vehicle center of gravity, l r is the distance from the rear axle to the vehicle center of gravity, δ SAS is the tire rotation angle, is the estimated value of the rate of change of speed in the XYZ directions, is the estimated value of the vehicle speed V XY 's estimated value, is the estimated value of the yaw rate ψ, is the estimated value of the sideslip angle β, is the estimated value of the yaw angular velocity, and T is the discrete time.

[0123] Furthermore, the discrete process model of the dynamic model is implemented as follows, as shown in Equation (7):

[0124]

[0125] Among them, m is the vehicle mass, g is the gravitational potential energy, I z is the yaw inertia moment, C f is the cornering stiffness of the front tire, C r is the cornering stiffness of the rear tire.

[0126] Furthermore, the method further includes the following steps:

[0127] Step S201, update the probability of the j-th candidate model through the likelihood function of the j-th candidate model to obtain the updated model probability;

[0128] Step S202: According to the updated model probabilities, perform a weighted sum of the estimation results of the filters of all interactive multi-models to calculate the final state estimation result;

[0129] Step S203: According to the updated model probabilities and the final state estimation result, perform a weighted sum of the estimation results of the filters of all interactive multi-models to calculate the final covariance estimation result.

[0130] Specifically, the measurement vector of each filter includes the data of the yaw rate sensor and GPS. The filter design process based on the interactive multi-model is as follows:

[0131] Update the model probabilities:

[0132] Use the likelihood function to update the model probabilities. The likelihood function of the j-th candidate model is shown in Equation (8):

[0133]

[0134] where v j (k) is the innovation matrix, S j (k) is the innovation covariance matrix, is the inverse matrix of the innovation covariance matrix, is the transpose of the innovation matrix, Λ j (k) is the likelihood function of the j-th candidate model. Assuming j = 1, 2, then when j = 1, it represents the kinematic model; when j = 2, it represents the dynamic model.

[0135] The probability update of the j-th candidate model is shown in Equation (9):

[0136]

[0137] where c is the normalization constant, and Λ j (k) is the likelihood function of the j-th candidate model, is the predicted probability of the j-th candidate model, μ j (k) is the probability of the j-th candidate model.

[0138] According to the updated model probabilities, perform a weighted sum of the estimation results of the filters of all interactive multi-models to calculate the final state estimation result As shown in Equation (10):

[0139]

[0140] where μ j (k) is the probability of the j-th candidate model, Is the result of the state estimation of the cubature Kalman filter, Is the final state estimation result;

[0141] According to the updated model probability and the final state estimation result, the estimation results of all the filters of the interacting multiple models are weighted and summed to calculate the final covariance estimation result, as shown in Equation (11):

[0142]

[0143] where μ j (k) is the probability of the j-th candidate model, Is the result of the state estimation of the cubature Kalman filter, P j (k|k) is the state covariance estimation result of the cubature Kalman filter, P(k|k) is the final covariance estimation result, Is the final state estimation result.

[0144] The position information (X, Y) of the vehicle can be estimated by fusing the following information, including: the vehicle position information (X GPS , Y GPS ) in the geodetic coordinate system obtained based on GPS, and the output vehicle speed V XY , yaw rate ψ, and sideslip angle β from the longitudinal vehicle state estimation module and the lateral vehicle state estimation module.

[0145] Furthermore, the prediction function of the observer is implemented in the following manner, as shown in Equation (12):

[0146]

[0147] where f(·) is the system nonlinear equation, T is the discrete time, and Are the estimated values of the vehicle position information, Is the estimated value of the vehicle speed V XY , Is the estimated value of the yaw rate ψ, Is the estimated value of the sideslip angle β, P k∣k-1 , P k-1∣k-1 Are the covariances, Q k-1 Is the system process noise.

[0148] Using the GPS position data (correcting the predicted position data), as shown in Equation (13):

[0149]

[0150] where R k Is the measurement noise covariance matrix; K k Is the Kalman gain for updating the state estimation; Pk∣k-1 is the estimated error covariance matrix after time update but before measurement update; H k is the measurement matrix that maps variables in the state space to the measurement space; is the state estimate after measurement update; is the state prediction after time update; z k is the measurement value vector, and the measurement value vector includes GPS measurement data; I is the identity matrix; P k∣k is the estimated error covariance matrix after measurement update; X GPS 、Y GPS are the X and Y coordinates measured by GPS respectively.

[0151] Compared with the prior art, the beneficial effects of this application are:

[0152] 1) The accuracy and reliability of state estimation are improved. By fusing the data of GPS and various vehicle-mounted sensors, this application comprehensively utilizes the advantages of each sensor to make up for the deficiencies of a single sensor, thereby improving the accuracy and reliability of vehicle state estimation. Especially when the GPS signal is limited or fails, relying on the auxiliary data of vehicle-mounted sensors, this application can still provide relatively accurate state estimation, effectively avoiding the uncertainty brought by a single information source.

[0153] 2) The adaptive ability of the system is enhanced. This application adopts adaptive filtering and machine learning algorithms, which can adjust the parameters of the estimation model in real time according to different driving environments and working conditions. This adaptive ability enables the system to flexibly cope with changing roads and driving conditions, ensuring efficient and accurate state estimation in various complex scenarios, and improving the robustness and stability of the system.

[0154] 3) The computational efficiency and real-time performance are optimized. This application introduces edge computing technology to complete some computational tasks locally on the vehicle terminal, reducing data transmission latency and dependence on the network. Combined with efficient filtering algorithms and optimized computational architectures, the real-time performance of state estimation is significantly improved, ensuring that the vehicle can quickly obtain accurate state information when driving at high speed and quickly changing working conditions, providing timely support for vehicle safety and control.

[0155] Embodiment 2:

[0156] Figure 4 is the switching state estimation system of the adaptive vehicle provided by the second embodiment of the present invention.

[0157] Referring to Figure 4 , the system includes:

[0158] An acquisition module, configured to acquire vehicle acceleration through a longitudinal acceleration sensor and acquire wheel speed through a wheel speed sensor;

[0159] The first road slope calculation module is used to obtain the first road slope by applying vehicle acceleration and wheel speed to kinematic equations;

[0160] The second road slope calculation module is used to obtain the GPS speed through a GPS data receiver and calculate the second road slope based on the GPS speed;

[0161] The construction module is used to construct a prediction process model;

[0162] The update module is used to update the prediction process model with the first road slope and the second road slope to obtain the road slope at the current moment;

[0163] The compensation module is used to compensate for the vertical speed on the sloped road surface with the road slope at the current moment;

[0164] The acquisition module is used to acquire the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model;

[0165] The determination module is used to determine the vehicle speed, yaw rate, and sideslip angle based on the kinematic model or the dynamic model;

[0166] The design module is used to design the prediction function of the observer based on the vehicle speed, yaw rate, and sideslip angle;

[0167] The calibration module is used to calibrate the prediction function of the observer with the vehicle position information obtained by the GPS module to obtain the vehicle longitudinal and lateral position information and the vehicle body state information;

[0168] Among them, the vehicle body state information includes the calibrated vehicle speed, the calibrated yaw rate, and the calibrated sideslip angle.

[0169] Embodiment III:

[0170] Figure 5 It is a schematic structural diagram of the electronic device provided in Embodiment III of the present invention.

[0171] Referring to Figure 5 , the electronic device 100 includes one or more processors 102, one or more storage devices 104, an input device 106, an output device 108, and an image acquisition device 110. These components are interconnected through a bus system 112 and / or other forms of connection mechanisms (not shown). It should be noted that Figure 5 The components and structures of the illustrated electronic device 100 are exemplary and not restrictive. According to requirements, the electronic device may also have other components and structures.

[0172] The processor 102 can be implemented in at least one hardware form of a digital signal processor (DSP), a field programmable gate array (FPGA), or a programmable logic array (PLA). The processor 102 can be a central processing unit (CPU) or a combination of one or more of other forms of processing units with data processing capabilities and / or instruction execution capabilities, and can control other components in the electronic device 100 to perform desired functions.

[0173] The storage device 104 can include one or more computer program products, and the computer program products can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory can include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory can include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions can be stored on the computer-readable storage media, and the processor 102 can run the program instructions to implement the client functions (implemented by the processor) in the embodiments of the present invention described below and / or other desired functions. Various application programs and various data can also be stored in the computer-readable storage media, such as various data used and / or generated by the application programs, etc.

[0174] The input device 106 can be a device used by a user to input instructions, and can include one or more of a keyboard, a mouse, a microphone, a touch screen, etc.

[0175] The output device 108 can output various information (such as images or sounds) to the outside (for example, to the user), and can include one or more of a display, a speaker, etc.

[0176] The image acquisition device 110 can capture images desired by the user (such as photos, videos, etc.), and store the captured images in the storage device 104 for use by other components.

[0177] An embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the adaptive vehicle switching state estimation method provided in the above embodiment are implemented.

[0178] An embodiment of the present invention also provides a computer-readable medium having non-volatile program code executable by a processor. A computer program is stored on the computer-readable medium, and when the computer program is run by the processor, the steps of the adaptive vehicle switching state estimation method in the above embodiment are executed.

[0179] The computer program product provided by the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, reference can be made to the method embodiments and will not be elaborated herein.

[0180] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0181] In addition, in the description of the embodiments of the present invention, unless otherwise clearly defined and limited, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0182] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., which can store program code.

[0183] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and should not be construed as indicating or implying relative importance.

[0184] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any technician familiar with the technical field of the present invention can still modify the technical solutions described in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims described.

Claims

1. A method for estimating the switching state of an adaptive vehicle, characterized in that, The method includes: Collecting vehicle acceleration through a longitudinal acceleration sensor and collecting wheel speed through a wheel speed sensor; Obtaining a first road slope by using the vehicle acceleration and the wheel speed through a kinematic equation; Obtaining GPS speed through a GPS data receiver and calculating a second road slope according to the GPS speed; Constructing a prediction process model; Updating the prediction process model by using the first road slope and the second road slope to obtain the road slope at the current moment; Compensating the vertical speed on a sloped road surface by using the road slope at the current moment; Obtaining the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model; Determining vehicle speed, yaw rate and sideslip angle according to the kinematic model or the dynamic model; Designing a prediction function of an observer according to the vehicle speed, the yaw rate and the sideslip angle; Correcting the prediction function of the observer by using the vehicle position information obtained by a GPS module to obtain vehicle longitudinal and lateral position information and vehicle body state information; Wherein, the vehicle body state information includes corrected vehicle speed, corrected yaw rate and corrected sideslip angle.

2. The method for estimating the switching state of an adaptive vehicle according to claim 1, characterized in that The method further includes: Updating the probability of the j-th candidate model by using the likelihood function of the j-th candidate model to obtain an updated model probability; Performing weighted summation on the estimation results of filters of all interactive multiple models according to the updated model probability to calculate a final state estimation result; Performing weighted summation on the estimation results of filters of all the interactive multiple models according to the updated model probability and the final state estimation result to calculate a final covariance estimation result.

3. The method for estimating the switching state of an adaptive vehicle according to claim 1, wherein The kinematic equation is implemented in the following manner: where a accel is the vehicle acceleration, is the time derivative of the velocity, representing the rate of change of velocity in the XYZ directions, g is the gravitational acceleration, and sinθ is the sine value of the first road gradient.

4. The method for estimating the switching state of an adaptive vehicle according to claim 3, wherein Obtaining a first road slope by using the vehicle acceleration and the wheel speed through a kinematic equation includes: Calculating the first road slope according to the following formula: where θ is the first road slope, and V whl is the wheel speed, which is measured by the wheel speed collected by the wheel speed sensor. XYZ ​ 5. The method for estimating the switching state of an adaptive vehicle according to claim 1, characterized in that Calculating a second road slope according to the GPS speed includes: Calculating the second road slope according to the following formula: where θ′ is the second road slope, and tan -1 (·) is the arctangent function, is the GPS velocity component in the Z direction, is the GPS velocity component in the X direction, is the GPS velocity component in the Y direction.

6. The method for estimating the switching state of an adaptive vehicle according to claim 1, wherein Constructing a prediction process model includes: wherein, is the predicted state vector at time k based on the predicted state vector at time k-1, is the estimated value of the acceleration in the XYZ directions, is the estimated value of the rate of change of velocity in the XYZ directions, T is the discrete time, is the estimated value of the second road slope, sin -1 (·) is a non-sine function, cos(·) is a cosine function.

7. The method for estimating the switching state of an adaptive vehicle according to claim 1, wherein The discrete process model of the kinematic model is implemented in the following manner: where, l f is the distance from the front axle to the vehicle's center of gravity, l r is the distance from the rear axle to the vehicle's center of gravity, δ SAS is the tire steering angle, is the estimated value of the rate of change of speed in the XYZ directions, is the estimated value of the vehicle speed V XY and is the estimated value of the yaw rate ψ, is the estimated value of the sideslip angle β, is the estimated value of the yaw angular velocity, and T is the discrete time.

8. The switching state estimation method for an adaptive vehicle according to claim 1, wherein The discrete process model of the dynamic model is implemented in the following manner: Among them, m is the vehicle mass, g is the gravitational potential energy, and I z is the yaw moment of inertia, C f is the cornering stiffness of the front tires, and C r is the cornering stiffness of the rear tires.

9. The method for estimating the switching state of an adaptive vehicle according to claim 1, wherein The prediction function of the observer is implemented in the following manner: P k∣k-1 = F k-1 P k-1∣k-1 (F k-1 ) T + Q k-1 where f(·) is the system nonlinear equation, T is the discrete time, and is the estimated value of the vehicle position information, is the estimated value of the vehicle speed V XY ; is the estimated value of the yaw rate ψ, is the estimated value of the sideslip angle β, P k∣k-1 、P k-1∣k-1 is the covariance, Q k-1 is the system process noise.

10. An adaptive vehicle switching state estimation system, characterized in that, The system includes: A collection module, configured to collect vehicle acceleration through a longitudinal acceleration sensor and collect wheel speed through a wheel speed sensor; A first road slope calculation module, configured to obtain a first road slope by using the vehicle acceleration and the wheel speed through a kinematic equation; A second road slope calculation module, configured to obtain GPS speed through a GPS data receiver and calculate a second road slope according to the GPS speed; A construction module, configured to construct a prediction process model; An update module, configured to update the prediction process model by using the first road slope and the second road slope to obtain the road slope at the current moment; A compensation module, configured to compensate the vertical speed on a sloped road surface by using the road slope at the current moment; An acquisition module, configured to acquire the j-th candidate model, where the j-th candidate model is a kinematic model or a dynamic model; A determination module, configured to determine a vehicle speed, a yaw rate, and a sideslip angle according to the kinematic model or the dynamic model; A design module, configured to design a prediction function of an observer according to the vehicle speed, the yaw rate, and the sideslip angle; A calibration module, configured to calibrate the prediction function of the observer by using vehicle position information obtained by a GPS module to obtain vehicle longitudinal and lateral position information and vehicle body state information; Wherein, the vehicle body state information includes a calibrated vehicle speed, a calibrated yaw rate, and a calibrated sideslip angle.

11. An electronic device, comprising a memory and a processor, wherein a computer program capable of running on the processor is stored on the memory, characterized in that, When the processor executes the computer program, the switching state estimation method of the adaptive vehicle according to any one of claims 1 to 9 above is implemented.

12. A computer-readable medium having non-volatile program code executable by a processor, characterized in that, The program code causes the processor to execute the switching state estimation method of the adaptive vehicle according to any one of claims 1 to 9.

Images