Automobile intelligent key positioning method, device and system and storage medium

Abstract

The invention discloses an automobile intelligent key positioning method, device and system and a storage medium, a built-in sensor of an intelligent key receives a field intensity signal emitted by a body control module BCM through a low-frequency antenna, a set membership filtering algorithm is adopted to carry out optimization processing on sensor data, and then an ellipsoid confidence region containing a real field intensity value is constructed; therefore, the robustness of key position judgment is obviously improved. By the adoption of the technical scheme, the problem that in the prior art, key positioning based on Bluetooth signal strength or a traditional filtering algorithm is prone to being affected by signal jitter and instantaneous interference, and consequently judgment inside or outside a vehicle is inaccurate is solved.

Description

Technical Field

[0001] This invention belongs to the field of automotive smart key technology, specifically relating to a method, device, system, and storage medium for locating automotive smart keys. Background Technology

[0002] The reliability and security of Passive Entry Passive Start (PEPS) systems fundamentally depend on the accurate sensing of the smart key's location, especially the accurate determination of the critical boundary between the inside and outside of the vehicle. Existing technologies to achieve this positioning function mainly follow the following technical routes, but all have inherent drawbacks: The threshold comparison method based on a single instantaneous measurement has a significant drawback: its extreme fragility. Any instantaneous electromagnetic interference or slight measurement noise can cause a single RSSI measurement to deviate significantly from the true value, leading to misjudgment. Due to its lack of interference immunity, this method cannot meet the high reliability and safety requirements of modern automobiles.

[0003] Data smoothing methods based on classical stochastic filtering algorithms have limitations because the assumptions underlying their theory are often difficult to meet in practical applications, specifically their dependence on noise statistical characteristics. For example, the optimality of Kalman filtering heavily relies on accurate knowledge of noise statistical characteristics (mean, variance). In the complex electromagnetic environment of automobiles, the statistical characteristics of noise may be time-varying, non-Gaussian, or even have heavy-tailed characteristics, meaning that the probability of large-amplitude impulse interference is much higher than that of a Gaussian distribution, which will lead to a sharp decline in filtering performance.

[0004] In summary, whether it is the basic threshold comparison method or the smoothing method based on classical random filtering, the core problem is that neither of them can fundamentally solve the problem of how to ensure the robustness of state estimation in environments with strong uncertainty, especially in the presence of bounded but potentially large pulse interference (such as relay attacks and noise from high-power charging piles). Summary of the Invention

[0005] To address the problems existing in the prior art, the present invention provides a method, device, system, and storage medium for locating a smart car key.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for locating a car smart key, comprising: The smart key's built-in sensors receive field strength signals transmitted by the Body Control Module (BCM) via a low-frequency antenna. By using an observer with historical data caching and mode switching capabilities, the field strength signal is optimized using a ensemble filtering algorithm to obtain an ellipsoidal confidence region containing the true field strength value.

[0007] Preferably, the original field strength value is filtered using a cluster filtering algorithm to obtain a field strength estimate; the field strength estimate is then compared with a preset threshold to determine the location of the smart key.

[0008] As a preferred option, if the current data is determined to be reliable, the observer uses a cluster filtering algorithm to fuse the current field strength observation with the recent valid historical data in the buffer and outputs an optimal field strength estimate. If the current data is detected to be abnormal or lost due to strong interference or attack, the observer discards the unreliable current data and relies only on the normal historical data sequence to drive the set member filtering algorithm for extrapolation estimation.

[0009] The present invention also provides a car smart key positioning device, comprising: The first processing module is used to enable the sensors built into the smart key to receive the field strength signal transmitted by the body control module (BCM) through a low-frequency antenna. The first processing module is used to optimize the field strength signal by using an observer with historical data caching and mode switching functions and a ensemble filtering algorithm to obtain an ellipsoidal confidence region containing the true field strength value.

[0010] Preferably, the second processing module uses a cluster filtering algorithm to filter the original field strength value to obtain a field strength estimate; the field strength estimate is then compared with a preset threshold to determine the location of the smart key.

[0011] Preferably, the second processing module is used to, if the current data is determined to be reliable, the observer uses a cluster filtering algorithm to fuse the current field strength observation value with the recent valid historical data in the buffer, and outputs an optimal field strength estimate. The second processing module is used to discard unreliable current data and drive the set member filtering algorithm to perform extrapolation estimation based solely on normal historical data sequences if the current data is detected to be abnormal or lost due to strong interference or attack.

[0012] The present invention also provides a car smart key positioning system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a car smart key positioning method when executed by the processor.

[0013] The present invention also provides a storage medium storing a computer program, which executes a car smart key positioning method when running.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: The smart key of this invention has a built-in sensor that receives the field strength signal transmitted by the body control module (BCM) through a low-frequency antenna. The sensor data is optimized using a ensemble filtering algorithm to construct an ellipsoidal confidence region containing the true field strength value, thereby significantly improving the robustness of key position determination.

[0015] To address the technical requirements of sensor data filtering, this invention proposes a set-membership filtering algorithm. By constructing a state feasible set instead of a traditional probability distribution description, it effectively avoids the requirement for precise modeling of noise statistical characteristics, instead utilizing its boundedness features as prior constraints, which is more in line with practical engineering applications. Furthermore, the observer proposed in this invention has a memory function, enabling it to continue providing meaningful estimates using recent reliable historical data when current sensor data is unavailable due to interference, thereby significantly improving the system's availability and robustness. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of the car smart key positioning method according to an embodiment of the present invention; Figure 2 This is a block diagram of the internal structure of the observer; Figure 3 For the ellipsoid in the membership filtering algorithm Iteration diagram. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Example 1

[0020] like Figure 1 As shown, the present invention provides a method for locating a car smart key, comprising: The first step, system modeling, involves the communication and measurement system between the smart key and the body control module (BCM), which can be described by the following discrete-time state-space equations:

[0021] in, Representing the Each observed object (e.g., corresponding to different low-frequency antenna channels) at time... The state vector is the true field strength value (RSSI) that needs to be estimated; At any moment For the first Each sensor measurement value; It is a nonlinear function; This is the initial value of the state.

[0022] , Let represent unknown but bounded process noise and detection noise, respectively, and satisfy the following ellipsoidal bounded constraints:

[0023] in, and It is a known positive definite weight matrix that defines the noise limit.

[0024] In this invention, noise is measured. , Mainly refers to the first Interference introduced when measuring field strength using a low-frequency antenna channel (such as electromagnetic noise from charging piles, relay attacks). The set-membership filtering algorithm constructs an ellipsoidal feasible set constrained by noise boundaries. Its mathematical essence lies in modeling unknown bounded interference as a convex polyhedral constrained space. This algorithm does not require precise knowledge of the statistical characteristics of the interference; it only needs to obtain the upper bound of the noise amplitude. Within the ellipsoidal domain jointly defined by the measurement data and the noise boundary, it ensures the accessibility of the true field strength value, thereby significantly improving the robustness of the positioning system.

[0025] Step 2: Design a novel observer with the following state estimation update law:

[0026] in, yes The estimated field strength at time [time]. It is the index of the backtracking steps. It is the length of the cached historical data window, which essentially stores the past... The observation data at each time point is cached and used for the current estimate. It is the observer gain matrix to be designed, and its optimization objective is to ensure the boundedness of the estimation error and to minimize the volume of the final ellipsoid estimation set. It is a smart weighting factor. In the historical data compensation mode of the observer, this factor can be dynamically adjusted according to data quality: when a... When the time-to-time data is reliable, When data at a given moment is detected to be unreliable due to interference, This allows the observer to automatically reduce or even eliminate the influence of unreliable historical data, relying more on data from other reliable periods. The internal structure of the observer is as follows: Figure 2 As shown.

[0027] The set-membership filtering algorithm used in this invention achieves robust estimation of field strength values ​​through the following steps. Its core lies in recursively calculating a minimal feasible ellipsoid set containing the true state, wherein the ellipsoid... like Figure 3 As shown (from left to right: initial ellipsoid, ellipsoid at the next moment, and optimal ellipsoid), the specific details are as follows: Initialization: Set the initial estimate of the field strength state of the smart key and its uncertainty boundary, and define an initial ellipsoid set based on this, which is guaranteed to contain the true initial field strength value. .

[0028] State prediction: Based on the system model formula (1), the ellipsoid set at the next moment is predicted. This step aims to obtain a predicted ellipsoid set, which must contain all possible states after the system model evolves, effectively accommodating bounded process noise (such as slow changes caused by the normal movement of the key).

[0029] Measurement Update: The predicted ellipsoid set is corrected using the actual field strength observations at the current moment. The update process aims to find the smallest ellipsoid that can encompass the current observation data while simultaneously considering bounded measurement noise (such as electromagnetic interference), thereby obtaining an updated and more accurate set of state estimates, the center of which is the optimal estimate. This process ensures that at any given time, the actual field strength value lies within the ellipsoid set calculated by the algorithm, providing worst-case reliability assurance for in-vehicle / out-of-vehicle judgment.

[0030] Assuming an antenna The initial field strength is constrained in an ellipse As shown, and its specific expression is as follows:

[0031] According to the antenna Initial estimate of field strength The previous moment's estimate With the current moment Acquired sensor measurement values , can obtain the first The next moment prediction equation for the root antenna field strength value:

[0032] for Time of the first Each sensor for antenna field strength The estimated value is derived from If the conditions are met, Cholesky decomposition can be used, requiring only auxiliary parameters. If satisfied, the following auxiliary equation can be obtained:

[0033] Therefore:

[0034] The filter error estimate is defined as: The system of estimated filtering error values ​​can be obtained as follows:

[0035] in

[0036] Rearranging equation (6) yields:

[0037] Furthermore, from the precondition of the set members—that the external noise is unknown but bounded—we can obtain:

[0038] If a positive scalar exists If such that the following inequality holds, then Established, that is, at any time The value is included Inside.

[0039]

[0040] Using Schul complement theorem, we can obtain the linear matrix inequalities.

[0041] in,

[0042] First, for the ellipsoid Given the ellipsoid constraint matrix The initial value constraints at the initial time are satisfied. In this case, it can be proven by mathematical induction. This ensures that the state estimate at each time step after the iterative algorithm is within the boundary of the constrained ellipsoid.

[0043] Secondly, for the ellipsoid The size can be determined by the ellipsoid constraint matrix. The trace value is used to represent the ellipsoid, and to obtain the optimal ellipsoid, the following convex optimization formula is optimized: trace The trace of the constraint matrix By solving the linear matrix inequality (9) and the convex optimization formula Thus, the local optimal parameters of the filter can be solved. Therefore, the field strength of the antenna at any time can be estimated using the observer system equation (2).

[0044] The specific process of locating the car smart key of the present invention includes: Phase 1: Signal Sensing and Acquisition

[0045] Signal transmission: When the user triggers the PEPS system (such as by pressing the start button), the Body Control Module (BCM) controls multiple low-frequency antennas located in different positions inside the vehicle (such as the driver's cabin, passenger cabin, and trunk) to transmit 125kHz low-frequency positioning signals in sequence.

[0046] Signal reception and measurement: The sensors inside the smart key receive these low-frequency signals and measure their field strength. (i.e., RSSI value). The data measured at this point is the raw data that includes environmental noise and potential interference.

[0047] Phase Two: Intelligent Observation and Crowd Estimation Data reception and anomaly diagnosis: Raw field strength data is transmitted to a novel observer designed for the vehicle. The primary task of this observer is to perform real-time anomaly detection, detecting situations where sensor data cannot be received for extended periods or exceeds conventional physical constraints.

[0048] Dual-mode adaptive estimation: The observer intelligently switches its operating mode based on the diagnostic results, which is the key to achieving high robustness.

[0049] Mode 1: Normal Filtering Mode: If the current data is deemed reliable, the observer enters this mode. The ensemble filtering algorithm fuses the current field strength observation with recent valid historical data in the buffer to output an optimal field strength estimate. This mode fully utilizes the latest information, ensuring the system's real-time performance.

[0050] Mode 2: Historical Data Compensation Mode: If the current data is detected to be abnormal or lost due to strong interference (such as high-power charging piles) or attacks, the observer immediately switches to this mode. In this mode, the observer discards unreliable current data and relies solely on normal historical data sequences to drive the set-membership filtering algorithm for extrapolation estimation. This is equivalent to providing the system with a "memory" function, maintaining reliable output even in extreme cases, greatly improving the system's availability.

[0051] Phase Three: Decision Making Threshold comparison: Regardless of the mode used, the observer will eventually output an accurate field strength estimate. This estimate has filtered out transient disturbances and uncertainties to the greatest extent possible. The system then compares this estimate with a fixed field strength threshold preset through precise experiments. Comparison; where the threshold Generally, data needs to be obtained through data calibration. Essentially, it utilizes the principle that the closer the key is to the antenna, the greater the field strength value. Calibration is used to determine a critical value for the inside and outside of the vehicle.

[0052] Location determination: Make a final determination based on the comparison results.

[0053] The system indicates that the smart key is located inside the vehicle.

[0054] The system indicates that the smart key is located outside the vehicle.

[0055] Phase 4: System Execution Execution control: The PEPS main controller executes the corresponding commands based on the position determination result.

[0056] If the system determines that "the key is inside the vehicle", it sends a start permission command to the engine control system, allowing the user to start the vehicle normally.

[0057] If the system determines that "the key is outside the car", then starting the car is prohibited.

[0058] This invention has the following technical features: 1. Use the membership filtering algorithm to process the original field strength value. Filtering is performed to obtain the field strength estimate. The field strength estimate is... With preset threshold The location of the smart key is determined by comparison.

[0059] Technical Features: This invention abandons the traditional random noise assumption and instead adopts a bounded noise model. The set-membership filtering algorithm does not pursue a single "optimal" estimate, but recursively calculates a defined set that guarantees the true field strength value at any given time. Included.

[0060] Technical problem to be solved: The performance of traditional Kalman filtering relies heavily on prior knowledge of noise statistical characteristics, which is difficult to satisfy in complex real-world automotive environments, leading to inaccurate estimations and unstable judgments.

[0061] 2. Design a novel observer with historical data caching and mode switching functions to achieve operation under extreme conditions (such as strong external interference). , Continuous and reliable estimation of situations such as relay attacks causing abnormal data transmission.

[0062] Technical features: In historical data compensation mode, the observer discards unreliable real-time data. Relying solely on cached valid historical data sequences The driving ensemble filtering algorithm performs extrapolation estimation, thus enabling the acquisition of an accurate field strength estimate even after signal interruption. .

[0063] Technical problem solved: When faced with extreme electromagnetic interference (such as high-power charging piles) or malicious attacks that cause instantaneous signal loss or continuous distortion, traditional methods become completely ineffective because they rely solely on current data.

[0064] 3. Construct a complete PEPS location determination system based on robust estimation. Technical problem to be solved: Improve the overall reliability, security (prevention of false positives) and availability of the entire PEPS system in complex real-world environments. Example 2

[0065] The present invention also provides a car smart key positioning device, comprising: The first processing module is used to enable the sensors built into the smart key to receive the field strength signal transmitted by the body control module (BCM) through a low-frequency antenna. The first processing module is used to optimize the field strength signal by using an observer with historical data caching and mode switching functions and a ensemble filtering algorithm to obtain an ellipsoidal confidence region containing the true field strength value.

[0066] As one embodiment of the present invention, the second processing module uses a member filtering algorithm to filter the original field strength value to obtain a field strength estimate; the field strength estimate is compared with a preset threshold to determine the location of the smart key.

[0067] As one embodiment of the present invention, the second processing module is used to, if the current data is determined to be reliable, the observer uses a set member filtering algorithm to fuse the current field strength observation value with the recent valid historical data in the buffer area and output an optimal field strength estimate. The second processing module is used to discard unreliable current data and drive the set member filtering algorithm to perform extrapolation estimation based solely on normal historical data sequences if the current data is detected to be abnormal or lost due to strong interference or attack. Example 3

[0068] The present invention also provides a car smart key positioning system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a car smart key positioning method when executed by the processor. Example 4

[0069] The present invention also provides a storage medium storing a computer program, which executes a car smart key positioning method when running.

[0070] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for locating a car smart key, characterized in that, include: The smart key's built-in sensors receive field strength signals transmitted by the Body Control Module (BCM) via a low-frequency antenna. By using an observer with historical data caching and mode switching capabilities, the field strength signal is optimized using a ensemble filtering algorithm to obtain an ellipsoidal confidence region containing the true field strength value.

2. The car smart key positioning method as described in claim 1, characterized in that, The original field strength value is filtered using a cluster filtering algorithm to obtain an estimated field strength value; the estimated field strength value is then compared with a preset threshold to determine the location of the smart key.

3. The car smart key positioning method as described in claim 2, characterized in that, If the current data is determined to be reliable, the observer uses a cluster filtering algorithm to fuse the current field strength observation with the recent valid historical data in the buffer and outputs an optimal field strength estimate. If the current data is detected to be abnormal or lost due to strong interference or attack, the observer discards the unreliable current data and relies only on the normal historical data sequence to drive the set member filtering algorithm for extrapolation estimation.

4. A car smart key positioning device, characterized in that, include: The first processing module is used to enable the sensors built into the smart key to receive the field strength signal transmitted by the body control module (BCM) through a low-frequency antenna. The first processing module is used to optimize the field strength signal by using an observer with historical data caching and mode switching functions and a ensemble filtering algorithm to obtain an ellipsoidal confidence region containing the true field strength value.

5. The automotive smart key positioning device as described in claim 4, characterized in that, The second processing module uses a cluster filtering algorithm to filter the original field strength value to obtain a field strength estimate; the field strength estimate is then compared with a preset threshold to determine the location of the smart key.

6. The automotive smart key positioning device as described in claim 5, characterized in that, The second processing module is used to, if the current data is determined to be reliable, the observer uses a cluster filtering algorithm to fuse the current field strength observation value with the recent valid historical data in the buffer area and output an optimal field strength estimate. The second processing module is used to discard unreliable current data and drive the set member filtering algorithm to perform extrapolation estimation based solely on normal historical data sequences if the current data is detected to be abnormal or lost due to strong interference or attack.

7. A car smart key positioning system, characterized in that, include: A memory and a processor, wherein the memory stores a computer program executed by the processor, the computer program performing the vehicle smart key location method as described in any one of claims 1-3 when executed by the processor.

8. A storage medium, characterized in that, The storage medium stores a computer program, which executes the car smart key positioning method as described in any one of claims 1-3 when it runs.

Images