A remote monitoring method and system for an electric vehicle range extending battery pack

Abstract

This invention relates to the field of electric vehicle technology and discloses a remote monitoring method and system for electric vehicle range extender battery packs. The method includes: acquiring first location information through a positioning module integrated on the range extender battery pack; automatically switching to a backup monitoring mode when a positioning signal loss is detected; and in the backup monitoring mode, performing dead reckoning using an inertial measurement unit and wheel speed sensors to generate second location information, and caching the second location information and battery status data in a local storage unit. This invention, by automatically switching to a backup monitoring mode when a positioning signal is lost, relying on an inertial measurement unit and wheel speed sensors to perform dead reckoning, and using a pre-stored lightweight digital map for matching and correction, can continuously provide effective location information in signal blind spots, solving the problems of positioning interruption and inability to continuously monitor in traditional solutions, and ensuring the continuity of monitoring.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle technology, specifically to a method and system for remote monitoring of electric vehicle range extender battery packs. Background Technology

[0002] Range-extended electric vehicles rely on range-extending battery packs to supplement their driving range, and some vehicles are provided to users through leasing or sharing models. Operators need to monitor the location and operating status of the battery packs in real time to achieve asset monitoring and anomaly alerts, ensuring asset security.

[0003] However, in real-world applications, vehicles often enter areas with weak or no GPS signals, such as underground parking lots, tunnels, and densely built-up areas. In these situations, the positioning module cannot output valid location information, and conventional remote monitoring solutions suffer from location interruptions and loss of connectivity, making it difficult to continuously track the battery pack's location. Furthermore, it cannot issue timely warnings in case of abnormal situations such as vehicles crossing boundaries or battery theft.

[0004] Some existing location-free monitoring solutions use inertial dead reckoning for point replacement, but inertial sensors have inherent drift errors. As the reckoning time increases, the position error accumulates rapidly, eventually deviating significantly from the actual position, failing to meet the accuracy requirements for long-term standby monitoring. Furthermore, after signal recovery, most existing solutions directly stitch the re-recovered dead position onto the original trajectory without considering error smoothing between the two trajectory segments, which can easily lead to trajectory jumps, affecting the continuity and accuracy of monitoring. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for remote monitoring of electric vehicle range extender battery packs, so as to solve the problems mentioned in the background art.

[0006] In a first aspect, the present invention provides: a method for remote monitoring of an electric vehicle range extender battery pack, comprising: The first location information is obtained by a positioning module integrated on the range-extended battery pack, and the first location information and battery status data are uploaded to the cloud platform through a cellular communication module. When a location signal loss is detected, the system automatically switches to backup monitoring mode. In the standby monitoring mode, dead reckoning is performed using an inertial measurement unit and wheel speed sensors to generate second position information, and the second position information and the battery status data are cached in a local storage unit. In standby monitoring mode, when the wheel speed sensor output is zero and the accelerometer output amplitude of the inertial measurement unit is equal to the gravitational acceleration, the vehicle is determined to be stationary, and the dead reckoning position error and speed error are simultaneously reset to zero, serving as the re-initialization point for subsequent dead reckoning. Using a lightweight digital map pre-stored in the local storage unit, the second location information obtained by dead reckoning is corrected by map matching. When the corrected location deviates from the feasible road area in the map, the second location information is forcibly corrected to the nearest feasible road. Once the positioning signal is restored, the cached second location information and battery status data are automatically transmitted to the cloud platform, which then stitches the trajectory together.

[0007] Furthermore, the lightweight digital map includes information on road direction, intersection shape, and road traffic direction; the map matching correction uses a particle filtering method to match the second location information obtained from dead reckoning with the road network in the map, and outputs the road location with the highest matching probability as the corrected location.

[0008] Furthermore, during dead reckoning, when the pulse interval of the wheel speed sensor is detected to increase and the horizontal acceleration component output by the accelerometer remains zero, it is determined that the vehicle has entered a state of uniform linear motion, and the acceleration integral weight in dead reckoning is reduced to suppress the accumulation of speed error.

[0009] Furthermore, the local storage unit pre-stores offline map tiles, which cover road and landmark information within a predetermined geographical area; when the map data required for map matching and correction exceeds the range of offline tiles, new map tiles are automatically requested to be downloaded during the first data retransmission after the positioning signal is restored.

[0010] Furthermore, when the cumulative driving distance calculated by the inertial measurement unit for dead reckoning exceeds the first threshold and no stationary state is detected during this period, an active error correction request is triggered. This reduces the reporting frequency of battery status data to lower cache pressure, and when the vehicle passes a known landmark location again, the dead reckoning result is corrected at a single point using the landmark location.

[0011] Furthermore, the known landmark locations include parking lot entrances and exits, gas stations, toll stations, or roadside units; when a vehicle passes a landmark location, the fixed position coordinates of the landmark device are read through short-range wireless communication, and these coordinates are used as the absolute reference point for dead reckoning to zero out the cumulative error.

[0012] Furthermore, the trajectory stitching step includes: after the positioning signal is restored, continuously receiving multiple restored GPS positioning points, and calculating the vehicle's heading and speed into the signal area based on these multiple positioning points; performing weighted smoothing on the end of the reverse-calculated trajectory and the beginning of the dead reckoning trajectory in the cache to minimize the position error variance of the stitching point; and temporarily increasing the position deviation trigger threshold of the electronic fence boundary crossing alarm for a predetermined period of time after the stitching is completed.

[0013] Furthermore, in the standby monitoring mode, the velocity vector sequence and corresponding timestamp during the dead reckoning process are cached simultaneously. When the positioning signal is restored, the velocity vector sequence is checked for consistency with the actual velocity vector sequence calculated from the GPS positioning point after retransmission. If the deviation between the two exceeds a preset ratio, it is determined that there is a sensor offset fault in the dead reckoning, and a sensor calibration reminder message is generated and pushed to the user terminal.

[0014] Furthermore, the inertial measurement unit includes a three-axis accelerometer and a three-axis gyroscope; the wheel speed sensor is the original wheel speed sensor of the vehicle, which is connected to the communication management unit of the range extender battery pack through the controller local area network bus; when the wheel speed sensor data is abnormal or lost, it automatically switches to using only the three-axis accelerometer and three-axis gyroscope for dead reckoning, and marks the position confidence as reduced.

[0015] Secondly, the present invention provides: a remote monitoring system for an electric vehicle range extender battery pack, used to execute the aforementioned remote monitoring method for an electric vehicle range extender battery pack, comprising: Intelligent range-extending battery pack; The positioning module, cellular communication module, inertial measurement unit, local storage unit, and communication management unit are integrated into the intelligent range-extending battery pack. The local storage unit pre-stores lightweight digital maps and offline map tiles; The communication management unit is used to monitor the validity of the positioning signal, control the positioning module and the cellular communication module to work when the signal is normal, and automatically switch to the backup monitoring mode when the positioning signal is lost. The inertial measurement unit and wheel speed sensor are used for dead reckoning in standby monitoring mode; The communication management unit is also used to determine that the vehicle is stationary when the output of the wheel speed sensor is zero and the amplitude of the accelerometer output of the inertial measurement unit is equal to the gravitational acceleration, and to set the dead reckoning position error and speed error to zero as a re-initialization point. The communication management unit is also used to perform map matching correction on the dead reckoning results using the lightweight digital map, and to cache the corrected second location information and battery status data in the local storage unit.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention automatically switches to backup monitoring mode when the positioning signal is lost, relies on inertial measurement unit and wheel speed sensor to complete dead reckoning, and uses pre-stored lightweight digital map for matching and correction. It can continuously provide effective position information in signal blind areas, solves the problems of positioning interruption and inability to continuously monitor in traditional solutions, and ensures the continuity of monitoring.

[0017] 2. This invention initializes the position and velocity errors of dead reckoning to zero when the vehicle is stationary, and reduces the weight of acceleration integral under uniform linear motion to suppress error accumulation. It can also use known landmark positions to perform single-point correction of accumulated errors, effectively controlling the accumulation of errors in inertial dead reckoning, improving the positioning accuracy in signal blind zones, and avoiding the problem of serious position deviation from reality after long-term reckoning.

[0018] 3. This invention performs weighted smoothing on the spliced ​​trajectory after signal recovery, which reduces the positional error of the splicing points, avoids trajectory jumps, and improves the continuity and accuracy of the overall trajectory. At the same time, in conjunction with sensor consistency verification, it can promptly detect sensor faults and remind calibration, ensuring the long-term reliability of the monitoring system. Attached Figure Description

[0019] Figure 1 This is a flowchart of a remote monitoring method for an electric vehicle range extender battery pack according to the present invention. Detailed Implementation

[0020] 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.

[0021] The present invention will now be described in further detail with reference to the embodiments and accompanying drawings.

[0022] Example 1: Please refer to Figure 1 This invention provides: a method for remote monitoring of an electric vehicle range extender battery pack, comprising: S100: Obtain first location information through a positioning module integrated on the range-extended battery pack, and upload the first location information and battery status data to the cloud platform through a cellular communication module; Specifically, when the vehicle is successfully started and in an open area, its positioning module (such as the u-blox NEO-M9N model that supports GPS, BeiDou and GLONASS, the three major global satellite navigation systems) will continuously calculate and output the vehicle's first position information at a stable frequency of once per second. During this process, the positioning module fully receives and processes navigation signals from multiple satellite systems to ensure accurate and reliable positioning data updates in open environments.

[0023] Cellular communication modules (such as the Quectel EC20 series which supports 4G full network connectivity) use the lightweight MQTT message transmission protocol as the communication standard. The specific workflow is as follows: the cellular communication module automatically collects and integrates the vehicle's first location information and its current battery status data at fixed 30-second intervals. Then, it packages this structured data into data packets that conform to the protocol format and uploads them to the preset cloud data processing and analysis platform through the established cellular network connection to realize remote monitoring and management functions.

[0024] S200: When the location signal is detected to be lost, it automatically switches to the backup monitoring mode; Specifically, the communication management unit continuously monitors the output of the positioning module, counting in real time the number of times it continuously outputs invalid positioning data (e.g., unable to obtain valid latitude and longitude coordinates or timestamp information). Once this number reaches a preset threshold N (usually set to 5 times), it means that valid location information has failed to be obtained in multiple consecutive attempts, and the system determines that the current positioning signal has been lost or is unavailable. Subsequently, the management unit immediately executes a preset energy-saving strategy, actively cutting off the regular power supply to the positioning module to reduce the overall power consumption of the system. At the same time, the system synchronously activates the backup monitoring mode, waking up the built-in inertial measurement unit through a specific command, enabling it to enter a continuous and stable working state. The inertial measurement unit then begins to collect motion parameters such as acceleration and angular velocity in real time, preparing sufficient data for subsequent position estimation using dead reckoning, thereby maintaining basic position tracking capability during periods of satellite signal loss.

[0025] S300. In the standby monitoring mode, dead reckoning is performed using an inertial measurement unit and wheel speed sensor to generate second position information, and the second position information and the battery status data are cached in the local storage unit. Specifically, the inertial measurement unit (IMU) integrates a three-axis accelerometer and a three-axis gyroscope, continuously monitoring the vehicle's motion in three-dimensional space and outputting high-frequency linear acceleration data along three orthogonal axes and angular velocity data along three axes in real time. Simultaneously, wheel speed sensors, via the vehicle's standard CAN bus interface, read the wheel speed pulse signals output by the original vehicle control system in real time. Combined with the vehicle's factory-preset or calibrated tire rolling radius parameters, the current instantaneous vehicle speed can be calculated. If the satellite positioning signal is lost, the dead reckoning algorithm immediately activates. Using the last valid first position coordinate before signal loss as the initial starting point, it performs a double integration of the acceleration data collected by the IMU to calculate the position offset relative to the starting point. Then, combined with the continuously measured changes in the vehicle's heading angle by the gyroscope, it recursively estimates the second position information for subsequent discrete moments. During dead reckoning, each time the system generates a set of second location data containing information such as position and speed, it is bound to key status monitoring data such as battery voltage, charging and discharging current, and internal temperature collected at the corresponding time. The data is written sequentially into a pre-opened circular cache queue in the local storage unit according to a strict timestamp order. This queue is responsible for temporarily storing all data during communication interruption and then retransmitting it to the remote server in batches after the subsequent communication link signal is restored.

[0026] S400. In standby monitoring mode, when the wheel speed sensor output is zero and the accelerometer output amplitude of the inertial measurement unit is equal to the gravitational acceleration, the vehicle is determined to be stationary, and the dead reckoning position error and speed error are simultaneously reset to zero, serving as the re-initialization point for subsequent dead reckoning. Specifically, when the wheel speed sensor detects no pulse signal output, it indicates that the wheels are completely stationary and not rotating. At the same time, by accurately calculating the three orthogonal acceleration components output by the triaxial accelerometer, and taking the square root of the sum of the squares of the three components, if the result is exactly equal to the standard gravitational acceleration g within the allowable error range, it fully proves that the vehicle, in its current state, is not subject to any additional horizontal or vertical acceleration components other than the Earth's gravity, meaning that the vehicle as a whole has not moved in any way. This confirms that the vehicle is absolutely stationary, completely eliminating any possibility of potential slippage or slow movement.

[0027] In this absolutely stationary state, since the vehicle has no displacement, the dead reckoning process based on inertial navigation principles will not generate any new effective position offset. However, the inherent zero-point drift and measurement errors of inertial sensors (such as gyroscopes and accelerometers) do not stop just because the vehicle is stationary. These errors will continue to accumulate during the integration calculation, causing the calculated position information to gradually deviate from the true position. Therefore, the precise physical position determined at the current moment will be directly set as the absolute position reference point. At this reference point, a crucial error reset operation will be performed: all position and velocity errors accumulated from the previous integration calculation will be directly cleared to zero, and all relevant parameters in the dead reckoning will be thoroughly reinitialized. Subsequent position calculations will start completely anew from this new, error-free absolute reference point. This fundamentally eliminates the errors that accumulate due to inertial sensor drift during long-term stationary monitoring. By periodically resetting the reference and initializing parameters at precise stationary moments, the overall navigation and positioning error generated during long-term standby monitoring can be effectively reduced to a minimum.

[0028] S500: Using a lightweight digital map pre-stored in the local storage unit, map matching correction is performed on the second location information obtained by dead reckoning. When the corrected location deviates from the feasible road area in the map, the second location information is forcibly corrected to the nearest feasible road. Specifically, the lightweight digital map pre-removes redundant geographic information elements such as building outlines, vegetation cover areas, and terrain elevation changes, while retaining the topology and key attributes of the road network, including the overall direction and curvature of roads, the geometric design of each intersection, the permitted traffic directions of each road, and the connectivity between different roads. After processing, the digital map's data volume is extremely compact, occupying very little local storage space and computing resources on terminal devices.

[0029] Map matching correction primarily relies on the particle filter algorithm. The particle filter algorithm first initializes a set of particles with different initial weights within the feasible road area surrounding the initial position (i.e., second position information) obtained from the dead reckoning module and its estimated error distribution range. Each particle represents a possible candidate position point on the road network. Then, the particle filter algorithm combines the vehicle's heading angle, speed, and other motion state information provided in real-time by dead reckoning to probabilistically evaluate the degree of match between each particle's position and the vehicle's actual trajectory. Based on the evaluation results, the particle weights are dynamically updated: particles that better match the vehicle's current heading and speed are assigned higher weights, while those that don't are assigned lower weights. Next, the particle filter algorithm performs a resampling step, eliminating invalid particles with too low a weight and negligible probability based on the weight distribution, while retaining and copying valid particles with high weights. After multiple rounds of iteration and filtering, the particle filter algorithm finally outputs the map road position corresponding to the particle with the highest matching probability among all particles, and uses this position as the new second position information after map matching correction.

[0030] The corrected location result will be further refined based on its relative relationship with the road network. If the corrected location falls within the feasible road area defined by the digital map, the result will be directly adopted, with only minor geometric adjustments to its coordinates to precisely align it with the nearest road centerline, conforming to the actual geometric representation of the road network. Conversely, if the corrected location still deviates from all existing feasible road areas, a projection repositioning strategy will be adopted: the location will be vertically projected onto the nearest feasible road segment, and the resulting coordinates will be output as the final corrected result. This effectively avoids the abnormal situation where the calculated location continuously shifts outside the roads due to the accumulation of dead reckoning errors.

[0031] When the positioning signal is restored, the S600 automatically transmits the cached second location information and battery status data to the cloud platform, and the cloud platform performs trajectory stitching.

[0032] Specifically, after the positioning module successfully outputs valid latitude and longitude coordinates N times consecutively, it determines that the positioning signal has been stably restored to normal working status. Then, it automatically executes the subsequent process. First, it shuts down the backup monitoring mode activated during the signal loss period and cuts off the power supply to the inertial measurement unit to effectively reduce the overall power consumption of the system. Then, it resumes the regular satellite positioning monitoring process to ensure that the main positioning function operates normally. At the same time, in the local storage unit, the circular cache queue has temporarily stored all the second location information with timestamps generated by the inertial measurement unit during the positioning signal loss period and the synchronously recorded battery status data. Once the communication management unit detects that the current cellular mobile communication signal is available, it will automatically start the data transmission program, batch package and integrate all data packets stored in the cache queue, and efficiently retransmit them to the remote cloud platform in one go through the cellular network.

[0033] After successfully receiving this batch of retransmitted data, the cloud platform will initiate a dedicated data fusion and trajectory reconstruction algorithm for processing. Specifically, the cloud platform will first perform time-sequential alignment and splicing of three segments of location data based on accurate timestamp information: the last valid first position recorded before the positioning signal was lost (i.e., satellite positioning position), all the retransmitted and corrected second positions (i.e., dead reckoning positions), and the first valid first position obtained after the positioning signal was recovered. During this process, the satellite positioning results of two key splicing points (i.e., the signal loss point and the signal recovery point) are used as position references. All second position coordinates located within the two splicing intervals are weighted and smoothed according to their spatial distance from the reference point. The processing principle is that the closer the position is to the reference point, the greater the weight is assigned, and the farther the position is from the reference point, the smaller the weight is assigned.

[0034] Furthermore, the lightweight digital map includes information on road direction, intersection shape, and road traffic direction; the map matching correction uses a particle filtering method to match the second location information obtained from dead reckoning with the road network in the map, and outputs the road location with the highest matching probability as the corrected location.

[0035] Specifically, the map matching correction process employs a particle filtering method. The specific implementation process is as follows: First, preliminary second location information is generated based on dead reckoning. Then, the particle filtering algorithm is used to perform multi-hypothesis probability matching between this location information and the road network constructed in the lightweight digital map. Through iterative calculations and weight updates, the matching degree of all possible paths is evaluated. Finally, the location of the road with the highest matching probability is selected from numerous candidate locations and output as the corrected optimal location result, thereby effectively improving positioning accuracy and reliability.

[0036] Furthermore, during dead reckoning, when the pulse interval of the wheel speed sensor is detected to increase and the horizontal acceleration component output by the accelerometer remains zero, it is determined that the vehicle has entered a state of uniform linear motion, and the acceleration integral weight in dead reckoning is reduced to suppress the accumulation of speed error.

[0037] Specifically, during vehicle positioning based on dead reckoning, multi-dimensional motion data from wheel speed sensors and accelerometers are continuously monitored. Once it is observed that the pulse signal intervals output by the wheel speed sensors gradually increase, and the acceleration components measured by the triaxial accelerometer in the horizontal direction (usually the vehicle's forward direction and the lateral direction perpendicular to it) remain stable near zero for a long time, it can be determined that the vehicle has entered an ideal state of uniform straight-line driving. To address the cumulative error in vehicle speed estimation that may be caused by acceleration noise and integral drift in this state, the data fusion strategy in dead reckoning will be dynamically adjusted. Specifically, the weighting coefficient for time integration of acceleration data will be significantly reduced, thereby effectively constraining the continuous growth of speed estimation error and improving the robustness and accuracy of overall positioning.

[0038] Furthermore, the local storage unit pre-stores offline map tiles, which cover road and landmark information within a predetermined geographical area; when the map data required for map matching and correction exceeds the range of offline tiles, new map tiles are automatically requested to be downloaded during the first data retransmission after the positioning signal is restored.

[0039] Specifically, the local storage unit pre-stores offline map tile data covering a specific geographical area. These tiles contain detailed and accurate information about the road network, traffic routes, and various key landmarks within the area. When the map data required for map matching and location correction exceeds the coverage of the currently stored offline tiles, a data update process will be automatically initiated. Specifically, during the first data synchronization process after a stable positioning signal is recaptured, a request will be proactively initiated to automatically download and supplement new map tiles for the corresponding missing areas from the server, thereby expanding the coverage of the offline map and ensuring the continuity and accuracy of subsequent positioning and navigation functions.

[0040] Furthermore, when the cumulative driving distance calculated by the inertial measurement unit for dead reckoning exceeds the first threshold and no stationary state is detected during this period, an active error correction request is triggered. This reduces the reporting frequency of battery status data to lower cache pressure, and when the vehicle passes a known landmark location again, the dead reckoning result is corrected at a single point using the landmark location.

[0041] Specifically, when the vehicle's travel distance calculated by the inertial measurement unit through dead reckoning exceeds a preset first threshold, if the vehicle is not detected to be stationary during this process, an active error correction request will be automatically triggered. In order to effectively alleviate the cache pressure of data processing and storage, the reporting frequency of battery status-related data will be reduced simultaneously. Subsequently, the system will wait for the vehicle to pass through a pre-known landmark location again during subsequent travel. Once the vehicle reaches the landmark, the system will use the precise location information of the landmark to perform a one-time single-point correction on the current location result calculated based on dead reckoning, thereby correcting the accumulated error and improving positioning accuracy.

[0042] Furthermore, the known landmark locations include parking lot entrances and exits, gas stations, toll stations, or roadside units; when a vehicle passes a landmark location, the fixed position coordinates of the landmark device are read through short-range wireless communication, and these coordinates are used as the absolute reference point for dead reckoning to zero out the cumulative error.

[0043] Specifically, the known landmark locations mentioned above include parking lot entrances and exits, gas stations, toll booths, and roadside units. These landmarks are not only important nodes in the road system but also reliable reference points for location positioning. When a vehicle passes through these significant locations, its communication system automatically identifies and reads the high-precision fixed coordinate information pre-stored in the landmark's device using efficient short-range wireless communication technology. This location coordinate, as a pre-measured and confirmed absolute spatial reference value, possesses high accuracy and stability. After obtaining these coordinates, the vehicle's navigation system immediately uses them as a reference to completely reset and zero out any accumulated errors that may exist in the dead reckoning based on inertial sensors. The vehicle can promptly and completely correct positioning deviations when passing through specific landmarks, thereby ensuring effective periodic correction and recovery of overall positioning accuracy and guaranteeing continuous and reliable navigation.

[0044] Furthermore, the trajectory stitching step includes: after the positioning signal is restored, continuously receiving multiple restored GPS positioning points, and calculating the vehicle's heading and speed into the signal area based on these multiple positioning points; performing weighted smoothing on the end of the reverse-calculated trajectory and the beginning of the dead reckoning trajectory in the cache to minimize the position error variance of the stitching point; and temporarily increasing the position deviation trigger threshold of the electronic fence boundary crossing alarm for a predetermined period of time after the stitching is completed.

[0045] Specifically, once the vehicle re-enters and acquires a valid positioning signal, it continuously receives and processes multiple GPS positioning point data transmitted since the signal recovery time. Based on this series of continuous positioning point information, a reverse calculation analysis method is used to deduce and calculate the vehicle's actual heading and real-time speed parameters when it first entered the signal coverage area. Subsequently, the end segment of the trajectory obtained from the reverse calculation is weighted and smoothed and fused with the beginning part of the trajectory segment previously generated and cached by dead reckoning (DR algorithm) during the signal loss period. The state at the splicing point is then finely adjusted to minimize the variance of the position error at that connection point globally, thereby achieving a smooth transition and accurate spatial geometric connection between the preceding and following trajectory segments.

[0046] In addition, to ensure stable operation and adaptability to possible state fluctuations in the short period of time after the trajectory stitching is completed, the position deviation trigger threshold used by the electronic fence boundary crossing alarm function is temporarily and appropriately increased within a pre-set time range after the stitching operation is completed, so as to avoid unnecessary false alarms caused by residual positioning errors that may exist in the early stage of trajectory stitching.

[0047] Furthermore, in the standby monitoring mode, the velocity vector sequence and corresponding timestamp during the dead reckoning process are cached simultaneously. When the positioning signal is restored, the velocity vector sequence is checked for consistency with the actual velocity vector sequence calculated from the GPS positioning point after retransmission. If the deviation between the two exceeds a preset ratio, it is determined that there is a sensor offset fault in the dead reckoning, and a sensor calibration reminder message is generated and pushed to the user terminal.

[0048] Specifically, throughout the entire operation of the standby monitoring mode, a series of continuous and ordered velocity vector data dynamically generated by the dead reckoning mechanism will be continuously collected and efficiently cached at a high frequency. Each independent velocity vector data point will be associated with a timestamp accurate to the millisecond level. Once the external high-precision positioning signal source (such as GPS) regains a stable and reliable connection from an interrupted state, the data fusion and verification process will be immediately initiated. This process will place the previously cached theoretically calculated velocity vector sequence and the actual velocity vector sequence, recalculated and interpolated based on the retransmitted historical GPS positioning data, within the same strictly aligned time interval frame for fine-grained comparison and trajectory consistency verification point by point. If, after rigorous verification and analysis, the results show that the velocity vector deviation between these two sequences at multiple consecutive corresponding points continuously and systematically exceeds the preset reasonable dynamic threshold ratio, it will be comprehensively determined that there is a high probability of cumulative inertial sensor offset or systematic error failure during the previous pure dead reckoning operation. Subsequently, the anomaly handling mechanism will be automatically triggered, generating a detailed calibration reminder report containing information such as the time range, deviation trend, and potentially affected sensors. This report will then be immediately pushed to the relevant monitoring terminal or maintenance personnel interface to clearly prompt users to perform timely manual inspection, data verification, and physical calibration of the relevant sensing devices, thereby ensuring the continuous reliability of subsequent positioning data and the safe and stable operation of the overall system.

[0049] Furthermore, the inertial measurement unit includes a three-axis accelerometer and a three-axis gyroscope; the wheel speed sensor is the original wheel speed sensor of the vehicle, which is connected to the communication management unit of the range extender battery pack through the controller local area network bus; when the wheel speed sensor data is abnormal or lost, it automatically switches to using only the three-axis accelerometer and three-axis gyroscope for dead reckoning, and marks the position confidence as reduced.

[0050] Specifically, the wheel speed sensor directly uses and integrates the original standard wheel speed sensor that comes standard with the vehicle at the factory. The wheel speed sensor establishes a two-way data communication transmission link with high stability and high reliability between itself and the communication management unit installed inside the vehicle's range extender battery pack housing through a common standardized controller area network bus communication protocol. In actual operation and monitoring, once the self-diagnostic mechanism detects abnormal continuous fluctuations or obvious abnormal jumps in the real-time vehicle speed data continuously fed back from the wheel speed sensor, or if the vehicle speed signal is completely interrupted and cannot be recovered due to a communication link failure, a preset, highly automated smooth switching process will be immediately initiated and executed to quickly switch to the backup emergency working mode. In this specific standby mode, the vehicle's core navigation and autonomous positioning functions will no longer rely on wheel speed signals. Instead, they will rely solely on the real-time linear acceleration data and three-dimensional angular velocity data provided by the high-precision three-axis accelerometer and three-axis gyroscope built into the inertial measurement unit. Through dead reckoning, the vehicle's relative position and trajectory will be continuously calculated and estimated, thus maintaining a basic and continuous position prediction and tracking capability even in abnormal situations.

[0051] In summary, this invention solves the problem of interrupted monitoring of the location and status of the range-extended battery pack in signal blind spots such as tunnels and underground parking garages by relying on the combination of inertial measurement unit and the vehicle's original sensors to perform dead reckoning when satellite positioning signals are lost, and completing position correction by combining pre-stored lightweight maps. With the help of a cache retransmission mechanism and cloud trajectory fusion reconstruction, it not only solves the problem of interrupted monitoring of the location and status of the range-extended battery pack, but also effectively suppresses the cumulative error of dead reckoning through a segmented dynamic error correction strategy. At the same time, it balances the power consumption and reliability of the monitoring system through scenario-based power consumption control and anomaly diagnosis mechanism, realizing continuous remote monitoring of the electric vehicle range-extended battery pack in all scenarios, and improving the stability of range-extended electric vehicle battery management and remote operation and maintenance.

[0052] Example 2: This invention provides: a remote monitoring system for an electric vehicle range extender battery pack, used to execute the remote monitoring method for an electric vehicle range extender battery pack described in Example 1, including: Intelligent range-extending battery pack; The positioning module, cellular communication module, inertial measurement unit, local storage unit, and communication management unit are integrated into the intelligent range-extending battery pack. The local storage unit pre-stores lightweight digital maps and offline map tiles; The communication management unit is used to monitor the validity of the positioning signal, control the positioning module and the cellular communication module to work when the signal is normal, and automatically switch to the backup monitoring mode when the positioning signal is lost. The inertial measurement unit and wheel speed sensor are used for dead reckoning in standby monitoring mode; The communication management unit is also used to determine that the vehicle is stationary when the output of the wheel speed sensor is zero and the amplitude of the accelerometer output of the inertial measurement unit is equal to the gravitational acceleration, and to set the dead reckoning position error and speed error to zero as a re-initialization point. The communication management unit is also used to perform map matching correction on the dead reckoning results using the lightweight digital map, and to cache the corrected second location information and battery status data in the local storage unit.

[0053] In summary, this system integrates positioning, communication, inertial sensing, and storage modules within the range-extended battery pack, reusing existing vehicle sensor data to achieve location completion. This eliminates the need for large-scale modifications to the vehicle's original architecture, ensuring continuous and stable monitoring of the range-extended battery pack's location and operational status even in scenarios with incomplete positioning signal coverage. This reduces modification costs and system power consumption. Furthermore, the dynamic error correction and fault diagnosis mechanisms effectively enhance the continuity and reliability of remote monitoring, meeting the battery operation and maintenance monitoring needs of range-extended electric vehicles across all scenarios.

[0054] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for remote monitoring of a range extender battery pack for electric vehicles, characterized in that, include: The first location information is obtained by a positioning module integrated on the range-extended battery pack, and the first location information and battery status data are uploaded to the cloud platform through a cellular communication module. When a loss of positioning signal is detected, the system automatically switches to backup monitoring mode. In the standby monitoring mode, dead reckoning is performed using an inertial measurement unit and wheel speed sensors to generate second position information, and the second position information and the battery status data are cached in a local storage unit. In standby monitoring mode, when the wheel speed sensor output is zero and the accelerometer output amplitude of the inertial measurement unit is equal to the gravitational acceleration, the vehicle is determined to be stationary, and the dead reckoning position error and speed error are simultaneously reset to zero, serving as the re-initialization point for subsequent dead reckoning. Using a lightweight digital map pre-stored in the local storage unit, the second location information obtained by dead reckoning is corrected by map matching. When the corrected location deviates from the feasible road area in the map, the second location information is forcibly corrected to the nearest feasible road. Once the positioning signal is restored, the cached second location information and battery status data are automatically transmitted to the cloud platform, which then stitches the trajectory together.

2. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: The lightweight digital map includes information on road direction, intersection shape, and road traffic direction. The map matching correction uses a particle filtering method to match the second location information obtained from dead reckoning with the road network in the map, and outputs the road location with the highest matching probability as the corrected location.

3. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: During dead reckoning, when the pulse interval of the wheel speed sensor increases and the horizontal acceleration component output by the accelerometer remains zero, it is determined that the vehicle has entered a state of uniform linear motion, and the acceleration integral weight in dead reckoning is reduced to suppress the accumulation of speed error.

4. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: The local storage unit pre-stores offline map tiles, which cover road and landmark information within a predetermined geographical area. When the map data required for map matching and correction exceeds the range of offline tiles, new map tiles are automatically requested to be downloaded during the first data retransmission after the positioning signal is restored.

5. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: When the cumulative driving distance calculated by the inertial measurement unit for dead reckoning exceeds the first threshold and no stationary state is detected during this period, an active error correction request is triggered. This reduces the reporting frequency of battery status data to lower cache pressure, and the dead reckoning results are corrected at a single point using the landmark location the next time the vehicle passes a known landmark location.

6. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 5, characterized in that: The known landmark locations include parking lot entrances and exits, gas stations, toll stations, or roadside units. When a vehicle passes a landmark location, the fixed position coordinates of the landmark device are read via short-range wireless communication, and these coordinates are used as the absolute reference point for dead reckoning to zero out the cumulative error.

7. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: The trajectory stitching step includes: after the positioning signal is restored, continuously receiving multiple restored GPS positioning points, and calculating the vehicle's heading and speed into the signal area based on these multiple positioning points; performing weighted smoothing on the end of the reverse-calculated trajectory and the beginning of the dead reckoning trajectory in the cache to minimize the position error variance of the stitching point; and temporarily increasing the position deviation trigger threshold of the electronic fence boundary crossing alarm for a predetermined period of time after the stitching is completed.

8. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: In standby monitoring mode, the velocity vector sequence and corresponding timestamp during dead reckoning are cached simultaneously. When the positioning signal is restored, the velocity vector sequence is checked for consistency with the actual velocity vector sequence calculated from the retransmitted GPS positioning point. If the deviation between the two exceeds a preset ratio, it is determined that there is a sensor offset fault in the dead reckoning, and a sensor calibration reminder message is generated and pushed to the user terminal.

9. The method for remote monitoring of an electric vehicle range extender battery pack according to claim 1, characterized in that: The inertial measurement unit includes a three-axis accelerometer and a three-axis gyroscope; the wheel speed sensor is the original wheel speed sensor of the vehicle, which is connected to the communication management unit of the range extender battery pack through the controller local area network bus; when the wheel speed sensor data is abnormal or lost, it automatically switches to using only the three-axis accelerometer and three-axis gyroscope for dead reckoning and marks the position confidence as reduced.

10. A remote monitoring system for an electric vehicle range extender battery pack, used to execute the remote monitoring method for an electric vehicle range extender battery pack according to any one of claims 1 to 9, characterized in that, include: Intelligent range-extending battery pack; The positioning module, cellular communication module, inertial measurement unit, local storage unit, and communication management unit are integrated into the intelligent range-extending battery pack. The local storage unit pre-stores lightweight digital maps and offline map tiles; The communication management unit is used to monitor the validity of the positioning signal, control the positioning module and the cellular communication module to work when the signal is normal, and automatically switch to the backup monitoring mode when the positioning signal is lost. The inertial measurement unit and wheel speed sensor are used for dead reckoning in standby monitoring mode; The communication management unit is also used to determine that the vehicle is stationary when the output of the wheel speed sensor is zero and the amplitude of the accelerometer output of the inertial measurement unit is equal to the gravitational acceleration, and to set the dead reckoning position error and speed error to zero as a re-initialization point. The communication management unit is also used to perform map matching correction on the dead reckoning results using the lightweight digital map, and to cache the corrected second location information and battery status data in the local storage unit.

Images