IMU-based vehicle driving speed dynamic monitoring method

Abstract

The invention relates to the technical field of vehicle driving monitoring, in particular to a vehicle driving speed dynamic monitoring method based on an IMU (Inertial Measurement Unit), which comprises the following steps: deploying an IMU sensor, synchronously acquiring vehicle acceleration and angular velocity data, performing multi-level filtering processing, analyzing the driving speed by utilizing a first time window, judging a frequent rapid deceleration state, and sending an early warning to a driver. The technical problems that under the complex road condition and the strong vibration environment, the sudden deceleration behavior of the vehicle cannot be accurately recognized, and misjudgment is likely to be generated are solved, the acceleration and angular velocity data of the vehicle are efficiently and synchronously collected through the IMU sensor, the sudden deceleration state is judged according to the ACCy value, and the vehicle speed is accurately detected. And meanwhile, the road surface flatness is judged in combination with the ACCz value, the rapid deceleration behavior of the vehicle is accurately recognized, the misjudgment rate is effectively reduced, and the rapid deceleration detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field related to vehicle driving monitoring, and in particular to a vehicle driving speed dynamic monitoring method based on IMU. Background Art

[0002] With the rapid development of autonomous driving and intelligent driving assistance systems, vehicle driving safety has gradually become an important issue. Traditional sudden deceleration detection technologies mostly rely on external sensors such as radar and cameras, but these sensors are expensive and easily affected by environmental factors (such as rainy and foggy weather and low light conditions), thus affecting system performance.

[0003] As a built-in sensor, IMU can provide real-time acceleration and angular velocity information of the vehicle and is less affected by external conditions such as weather and light, making it a more economical and real-time detection solution. However, in complex road conditions and strong vibration environments, IMU's detection results are prone to misjudgment, resulting in inaccurate judgment of sudden deceleration behavior.

[0004] In summary, the prior art has a technical problem that it is unable to accurately identify the sudden deceleration behavior of a vehicle under complex road conditions and strong vibration environments, which is prone to misjudgment. Summary of the invention

[0005] The present application provides an IMU-based vehicle driving speed dynamic monitoring method, aiming to solve the technical problem in the prior art that the sudden deceleration behavior of the vehicle cannot be accurately identified under complex road conditions and strong vibration environments, and misjudgment is easily caused.

[0006] In view of the above problems, the technical solution to implement this application is:

[0007] The present application provides a method for dynamic monitoring of vehicle driving speed based on an IMU, wherein the method comprises: deploying an IMU sensor based on a target vehicle, wherein a first sampling trigger signal corresponding to the IMU sensor and a second sampling trigger signal corresponding to the vehicle speed sensor are synchronous pulses; based on the first sampling trigger signal corresponding to the IMU sensor and the second sampling trigger signal corresponding to the vehicle speed sensor, collecting acceleration data and angular velocity data of the target vehicle during driving; performing multi-level filtering processing on the acceleration data and angular velocity data, and using a first time window to analyze the driving speed data; judging whether the target vehicle is in a state of frequent sudden deceleration through the driving speed data, and obtaining a judgment result; sending a warning message to the driver according to the judgment result, resetting the state, and preparing for the next monitoring.

[0008] In summary, one or more technical solutions provided in the present application solve the technical problem that the sudden deceleration behavior of the vehicle cannot be accurately identified and misjudgment is easy to occur under complex road conditions and strong vibration environments. It realizes the efficient and synchronous collection of the vehicle's acceleration and angular velocity data through the IMU sensor, and judges the sudden deceleration state according to the ACCy value. At the same time, the road surface flatness is judged in combination with the ACCz value, and the sudden deceleration behavior of the vehicle is accurately identified, which effectively reduces the misjudgment rate and improves the accuracy of sudden deceleration detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flow chart of a method for dynamic monitoring of vehicle driving speed based on IMU is provided for this application. DETAILED DESCRIPTION

[0010] Example

[0011] The present application is described in detail below with reference to the accompanying drawings. Figure 1 As shown, the present application provides a vehicle driving speed dynamic monitoring method based on IMU, wherein the method comprises:

[0012] S1: Based on the target vehicle, an IMU sensor is deployed, and the first sampling trigger signal corresponding to the IMU sensor and the second sampling trigger signal corresponding to the vehicle speed sensor are synchronous pulses; S2: Based on the first sampling trigger signal corresponding to the IMU sensor and the second sampling trigger signal corresponding to the vehicle speed sensor, the acceleration data and angular velocity data of the target vehicle during driving are collected.

[0013] Specifically, the target vehicle refers to the vehicle whose driving speed needs to be dynamically monitored, that is, the vehicle being detected; the IMU (Inertial Measurement Unit) sensor is a device that integrates sensors such as accelerometers and gyroscopes, and is used to measure the acceleration, angular velocity and spatial position changes of objects. IMU can provide continuous motion state data without relying on external signals, and is very suitable for real-time monitoring in dynamic environments.

[0014] The vehicle speed sensor is used to measure the vehicle's speed, usually based on wheel speed, GPS or radar technology, etc. The sensor usually provides real-time speed data of the vehicle; the first sampling trigger signal and the second sampling trigger signal refer to the synchronization signals of the IMU sensor and the vehicle speed sensor. The triggering mode of the two signals is the same, which is used to ensure that the vehicle's acceleration data, angular velocity data and vehicle speed data are collected at the same time. The two signals should be triggered synchronously to ensure that the data between the two are consistent in time.

[0015] IMU sensors are deployed on the target vehicle to measure the acceleration and angular velocity of the vehicle in real time, providing necessary motion data for subsequent driving behavior analysis; the vehicle speed sensor is used to provide the vehicle's driving speed data and provide the vehicle's motion status information at different time points.

[0016] The sampling signals of the IMU sensor and the vehicle speed sensor use a synchronous pulse method. The design of the synchronous pulse enables the two sensors to trigger data collection at the same time point. In this way, at the same moment, the IMU sensor and the vehicle speed sensor respectively record the acceleration, angular velocity and actual driving speed of the vehicle to ensure the time consistency between the data. This synchronization mechanism greatly improves the accuracy of data fusion and avoids the time difference problem caused by different sampling timings.

[0017] With the support of this synchronization mechanism, the IMU sensor starts to collect the vehicle's acceleration and angular velocity data through the first sampling trigger signal. The acceleration data reflects the vehicle's acceleration or deceleration process in the longitudinal and lateral directions, while the angular velocity data is used to reflect the vehicle's rotation or turning. At the same time, the vehicle speed sensor also collects the vehicle's instantaneous driving speed through the second sampling trigger signal. These data provide comprehensive dynamic information for subsequent driving behavior analysis.

[0018] In this way, by combining the data from the IMU and the vehicle speed sensor, the dynamic information of the target vehicle can be fully grasped, providing basic data support for subsequent driving status judgment, sudden deceleration detection, etc. For example, in the actual driving process, the combination of the vehicle's acceleration data and speed data can help determine whether the driver has performed sudden acceleration or deceleration operations, thereby providing a basis for subsequent driving behavior warning and analysis.

[0019] S3: Perform multi-level filtering on the acceleration data and angular velocity data, and use the first time window to analyze the driving speed data; S4: Determine whether the target vehicle is in a state of frequent sudden deceleration through the driving speed data, and obtain a judgment result; S5: Send a warning message to the driver based on the judgment result, reset the state, and prepare for the next monitoring.

[0020] Specifically, acceleration data refers to the acceleration of the vehicle, which refers to the rate of change of the vehicle's speed in different directions, including longitudinal (front and rear) acceleration and lateral (left and right) acceleration; angular velocity data refers to the rotation rate of the vehicle on three axes, which is often used to describe the rotation action that occurs when the vehicle turns or drives; multi-level filtering processing is to process the signal data layer by layer, and reduce noise by applying different filtering algorithms (such as low-pass filtering, Kalman filtering, etc.) to improve the accuracy and reliability of the signal.

[0021] In data analysis, a time window refers to a data sample collected and processed within a specific period of time. In vehicle monitoring, a time window can be used to analyze driving behavior over a period of time. A sudden deceleration state refers to a rapid deceleration of the vehicle in a short period of time. Usually, a sudden deceleration state will produce significant acceleration changes during driving, which may affect driving safety. A frequent sudden deceleration state refers to multiple sudden decelerations of the vehicle in a relatively short period of time, reflecting unstable driving habits or dangerous driving behaviors.

[0022] The collected acceleration and angular velocity data are subjected to multi-level filtering. The purpose of this step is to clean up the noise in the data and improve the accuracy and reliability of the data. Filtering can be done in a variety of ways, such as using a low-pass filter to remove high-frequency noise, or using a Kalman filter to more accurately estimate the vehicle state. These processes help to extract the vehicle's true motion trajectory and remove errors introduced by external factors or sensor instability.

[0023] The processed driving speed data is analyzed using the first time window. The first time window refers to the continuous collection and analysis of the vehicle's speed data within a specific time interval. For example, a window is set to 5 seconds, and the slide is 5 seconds each time. The data within these 5 seconds is analyzed to evaluate the vehicle's driving behavior during this period. By analyzing these data, it can be identified whether the vehicle has a sudden deceleration.

[0024] By analyzing the obtained driving speed data, it is determined whether the target vehicle has entered a state of frequent sudden deceleration, and based on the judgment result, a warning message is sent to the driver. For example, if it is detected that the vehicle is frequently decelerating suddenly, it is because the driver has braked suddenly, and a reminder is issued to let the driver pay attention to driving safety. After each judgment, the status is reset to prepare for the next round of monitoring. Through the above steps, real-time dynamic monitoring of the vehicle is achieved, the driver's sudden deceleration behavior is discovered in time and a warning is issued, thereby improving driving safety and reducing the potential risk of traffic accidents.

[0025] Furthermore, the present application method also includes:

[0026] The IMU sensor obtains the raw data of the six-axis gyroscope through the SPI interface; processes the raw data through low-pass filtering to obtain the first layer of processed data; and performs Kalman filtering based on the first layer of processed data to obtain the second layer of processed data.

[0027] Specifically, IMU sensors usually include components such as accelerometers, gyroscopes and magnetometers. IMU sensors can measure information such as acceleration, angular velocity and direction changes, and are used to dynamically monitor the motion state of objects; the SPI (Serial Peripheral Interface) interface supports high-speed data transmission and is used to read raw data from the six-axis gyroscope sensor.

[0028] A six-axis gyroscope is a sensor that can simultaneously measure the angular velocity and acceleration of an object in three-dimensional space. The six axes correspond to the object's three acceleration axes (x, y, and z axes) and three rotation axes (angular velocity around the x, y, and z axes). Low-pass filtering is used to remove high-frequency noise from the signal and retain only low-frequency components. Low-pass filtering reduces data errors caused by noise or irregular fluctuations. Kalman filtering is used to estimate and optimize noisy data to further improve data accuracy and eliminate high-frequency noise.

[0029] The IMU sensor is deployed in the vehicle to obtain the raw acceleration and angular velocity data from the six-axis gyroscope through the SPI interface. The six-axis gyroscope will provide the vehicle's acceleration (ACCx, ACCy, ACCz) and angular velocity (GyroX, GyroY, GyroZ) data in three axes in each sampling cycle. Through the SPI interface, these data will be transmitted to the processing unit (such as a microcontroller or computer) for subsequent processing.

[0030] The acquired raw data usually contains high-frequency components such as sensor noise and environmental interference, which will affect the accuracy of the data. Therefore, before processing the raw data, low-pass filtering is first used for preprocessing. Low-pass filtering can effectively remove high-frequency noise, making the low-frequency part of the signal (that is, the actual motion data of the vehicle) more prominent. After filtering, the first layer of processed data obtained contains more accurate acceleration and angular velocity information.

[0031] Although the low-pass filter has removed the high-frequency noise, there may still be some low-frequency errors or instabilities in the data. Therefore, the first layer of processed data is further optimized, and the estimated value is adjusted in real time based on the recursive algorithm to eliminate the deviation caused by the sensor error. The Kalman filter will use the previous data for prediction and adjust its estimate based on the new sampled data to generate the second layer of processed data. After Kalman filtering, it will be more accurate and reliable and better reflect the real dynamics of the vehicle. Through the above process, the raw data provided by the IMU sensor will be filtered layer by layer to obtain accurate acceleration and angular velocity data, which provides a basis for subsequent vehicle dynamic monitoring and driving behavior analysis.

[0032] Furthermore, the present application method also includes:

[0033] Based on the six-axis gyroscope, axial fusion is performed to determine the three-axis Euler angle; based on the three-axis Euler angle, two bidirectional linked lists are created, wherein the number of bits of each linked list is N, which are respectively used to store ACCy floating point numbers and ACCz floating point numbers.

[0034] Specifically, a six-axis gyroscope is a sensor that can measure acceleration and angular velocity at the same time. It usually includes a three-axis accelerometer and a three-axis gyroscope, which can measure the acceleration (ACCx, ACCy, ACCz) and angular velocity (GyroX, GyroY, GyroZ) of an object in three orthogonal directions respectively; axial fusion is the process of combining acceleration data and angular velocity data to obtain a more accurate direction and motion state. This process usually performs mathematical operations and transformations on the data in three directions (such as acceleration and angular velocity) to generate a result that is more in line with the actual motion trajectory. In this solution, axial fusion is used to infer Euler angles from data in three-dimensional space; Euler angles are a way to describe the posture of an object in three-dimensional space. The rotation state of an object is represented by three angles (roll angle, pitch angle, and yaw angle). For a vehicle, the three-axis Euler angle can represent the direction and posture of the vehicle relative to the ground in three-dimensional space.

[0035] A doubly linked list is a data structure in which each node contains pointers to the previous node and the next node. Each node can store two data elements (such as ACCy and ACCz). Through the doubly linked list, data insertion and deletion operations can be performed efficiently, especially when data needs to be frequently accessed and operated; ACCy and ACCz represent the acceleration data of the vehicle on the Y-axis and Z-axis respectively, stored in floating-point format.

[0036] The IMU sensor installed in the vehicle will provide data from the six-axis gyroscope, which includes acceleration in three directions (ACCx, ACCy, ACCz) and angular velocity in three directions (GyroX, GyroY, GyroZ). The raw data will be uploaded through the SPI interface or other communication protocols. Furthermore, in order to accurately describe the motion state and posture of the vehicle, the acceleration data (ACCx, ACCy, ACCz) and the angular velocity data (GyroX, GyroY, GyroZ) need to be axially fused. The purpose of axial fusion is to combine the information of the two to infer the orientation and dynamic behavior of the vehicle. Specifically, acceleration data (such as ACCx, ACCy, ACCz) is used to estimate the spatial direction of the vehicle, and angular velocity data is used to supplement and correct this direction estimate to obtain an accurate direction estimate. This process will generate three-axis Euler angles, representing the roll angle, pitch angle and yaw angle of the vehicle in space.

[0037] After axial fusion, the angle conversion algorithm is used to obtain the vehicle's three-axis Euler angles. The Euler angles describe the vehicle's spatial posture. For example, the roll angle represents the vehicle's rotation around the front axle, the pitch angle represents the vehicle's rotation around the horizontal axis, and the yaw angle represents the vehicle's rotation around the vertical axis. Through these three-axis Euler angles, the vehicle's direction in three-dimensional space can be accurately described.

[0038] Based on the vehicle's acceleration data (especially the acceleration data in the ACCy and ACCz directions), two bidirectional linked lists are created. Each linked list contains N nodes, and each node stores a pair of floating-point numbers: one is the ACCy floating-point number, which represents the acceleration of the vehicle in the Y-axis direction, and the other is the ACCz floating-point number, which represents the acceleration of the vehicle in the Z-axis direction. The bidirectional linked list can efficiently access and update data. The bidirectional nature of the linked list enables flexible forward and backward traversal of data during the processing process, which is suitable for scenarios where frequent adjustment, deletion or insertion of data is required.

[0039] When new data (such as new ACCy and ACCz acceleration data) is collected, it is added to the end of the bidirectional linked list. If the linked list has reached the preset length N, the oldest data node needs to be removed. This is achieved through the pointer of the bidirectional linked list. The new node is automatically linked to the end of the linked list, and the old node is deleted by adjusting the front and back pointers. Through the bidirectional linked list, the acceleration data is efficiently managed and updated during the continuous sampling process to ensure that the vehicle's acceleration data is in the latest state at all times.

[0040] In the above steps, by obtaining the raw data of the six-axis gyroscope, performing axial fusion to calculate the three-axis Euler angle, and combining a bidirectional linked list to store and update the acceleration data, the dynamic state of the target vehicle can be reflected efficiently and accurately, providing support for vehicle dynamic monitoring and driving behavior analysis.

[0041] Furthermore, the present application method also includes:

[0042] Each node contains two floating point numbers and pointers to the previous and next nodes to support bidirectional traversal; the two floating point numbers include a pair of an ACCy floating point number and an ACCz floating point number.

[0043] Specifically, a node refers to an element in a linked list, which is used to store data. In this method, each node contains two floating-point numbers and pointers to the previous and next nodes; ACCy floating-point number and ACCz floating-point number: represent the acceleration data of the target vehicle in the Y-axis and Z-axis directions, respectively, representing the acceleration magnitude of the vehicle in these directions; a bidirectional linked list is a linked list structure, in which each node contains not only data but also two pointers, pointing to the previous node and the next node respectively, and supports bidirectional traversal.

[0044] A bidirectional linked list will be used to store acceleration data. Each linked list node will contain two floating-point numbers: one is the acceleration value of the Y axis (ACCy), and the other is the acceleration value of the Z axis (ACCz). In addition, each node also contains two pointers, pointing to the previous and next nodes of the node, respectively, thus forming a bidirectional linked list structure. In this way, data can be traversed from both ends of the linked list to support flexible access and operation.

[0045] Each time new acceleration data (ACCy and ACCz) is obtained, a new node is created and the two floating-point numbers are stored in the node; the new node is inserted at the end of the linked list, and the pointers of the adjacent nodes in the linked list are updated so that the linked list remains bidirectionally connected, and the forward pointer of each node points to the previous node, and the backward pointer points to the next node.

[0046] By utilizing the characteristics of a bidirectional linked list, nodes can be traversed from beginning to end (forward) or from end to beginning (reverse). In this way, acceleration data can be flexibly accessed when data needs to be processed. If the number of nodes in the linked list exceeds the preset maximum value (N nodes), the oldest node (i.e., the node at the head of the linked list) will be deleted to keep the length of the linked list no longer than N, so as to ensure the real-time and accuracy of the data. In this way, acceleration data can be efficiently stored and accessed, providing support for subsequent processing and judgment.

[0047] Furthermore, the present application method also includes:

[0048] Based on the sampling frequency, a first time window is configured; based on the first time window, when data is collected, the currently collected ACCy value and ACCz value are added to the data set; at the same time, a sliding window mechanism is adopted to remove the original ACCy value and the original ACCz value in the data set each time new data is added.

[0049] Specifically, the ACCy floating point number and the ACCz floating point number represent the acceleration data of the vehicle in the Y-axis and Z-axis directions. ACCy represents the acceleration in the Y-axis direction, and ACCz represents the acceleration in the Z-axis direction. The first time window refers to a time period, which is used to limit the data collected within the time period. The sampling frequency determines the size of the window, that is, how many times the data is collected within the window. The sliding window mechanism means that in the data set, the window size is fixed, and as new data enters, the oldest data will be removed to keep the size of the data set unchanged.

[0050] The length of the first time window is determined according to the set sampling frequency. The sampling frequency determines the frequency of data collection. For example, data is collected once per second. According to this frequency, the time window will be defined as a fixed time period (such as 1 second, 10 seconds, etc.). Whenever the time window ends, the window will be updated and the collected data will also be updated to the data set.

[0051] During the collection process, when the vehicle's accelerometer (IMU sensor) obtains the acceleration data of ACCy and ACCz, these data will be added to the data set according to the current timestamp. Each collection of acceleration data includes a pair of floating numbers: an ACCy floating number (Y-axis acceleration) and an ACCz floating number (Z-axis acceleration). These data will be added to the set in chronological order.

[0052] To ensure the real-time and finiteness of the data set, the sliding window mechanism will work automatically when new data is added to the set. Specifically, every time a new data pair (ACCy, ACCz) is collected and added to the set, the oldest acceleration data in the set (i.e., the old values ​​of ACCy and ACCz) will be removed. In this way, the data in the data set always remains the latest set, and the length is always consistent, and memory overflow or low computing efficiency will not occur due to too much data.

[0053] Each time new data (ACCy, ACCz) is collected and added to the collection, it is updated according to the length of the time window and the oldest data is removed. In this way, all data in the data set are always the latest acceleration data collected within the first time window. By configuring the time window, collecting acceleration data and using the sliding window mechanism, the vehicle acceleration data can be efficiently managed and updated. Whenever new data comes in, the oldest data is removed, which reduces storage pressure and ensures the timeliness and accuracy of the data.

[0054] Furthermore, the present application method also includes:

[0055] The ACCy values ​​in the data set are averaged to obtain an average ACCy value; a sudden deceleration judgment threshold is set, and the average ACCy value is compared with the sudden deceleration judgment threshold; if the average ACCy value is less than the sudden deceleration judgment threshold, it is determined that the target vehicle is in a sudden deceleration state.

[0056] Specifically, the ACCy value represents the acceleration data of the vehicle in the Y-axis direction. The ACCy value is the original acceleration value obtained by the IMU sensor and is used to reflect the acceleration change of the vehicle in the Y-axis direction; the sudden deceleration judgment threshold is used to determine whether the vehicle has a sudden deceleration. The sudden deceleration judgment threshold is set through experiments or historical data analysis, representing a critical value of acceleration. When the acceleration is less than the sudden deceleration judgment threshold, it indicates that the vehicle has a sudden deceleration; the average ACCy value refers to the average of all the collected ACCy values ​​within a certain time window, and the average of the acceleration values ​​obtained represents the overall acceleration change of the vehicle in the Y-axis direction within the time period.

[0057] After collecting ACCy data within a time window, the data is averaged. Specifically, all ACCy values ​​within the time window are added together and then divided by the number of data points to obtain the average ACCy value. The average ACCy value can reflect the overall acceleration change of the vehicle within this time window and determine the movement trend of the vehicle.

[0058] According to the actual application scenario or historical data, a threshold for sudden deceleration judgment is set, which means that when the acceleration of the vehicle is lower than this value, the vehicle has experienced sudden deceleration, which is used as a standard for judging whether the vehicle has decelerated sharply. The calculated average ACCy value is compared with the preset sudden deceleration judgment threshold. If the average ACCy value is less than the sudden deceleration judgment threshold, it is considered that a sudden deceleration state of the target vehicle has been detected.

[0059] If the average ACCy value is less than the sudden deceleration judgment threshold, the target vehicle is judged to be in a sudden deceleration state, which means that the deceleration of the vehicle exceeds the preset sudden deceleration standard, which may be caused by emergency braking, excessive speed, etc.; once the vehicle is judged to be in a sudden deceleration state, the corresponding control mechanism, such as sending an alarm to the driver, recording event data, etc., reminds the driver to pay attention to driving safety. At the same time, the relevant state can be reset for the next monitoring to prepare for continued data collection. Through the above steps, the sharp deceleration of the vehicle is effectively monitored, and data support is provided for the subsequent control and alarm mechanisms, so as to reflect the dynamic state of the vehicle in real time and accurately, and enhance driving safety.

[0060] Furthermore, the present application method also includes:

[0061] After determining that the target vehicle is in a rapid deceleration state, the frequent rapid deceleration judgment process is started as follows:

[0062] A second time window and a counter corresponding to frequent sudden deceleration detection are set; within the second time window, the ACCy value is continuously monitored based on the driving speed data, and each time a sudden deceleration state is detected, the counter is incremented by 1.

[0063] Specifically, the second time window refers to the time interval used for monitoring in the frequent sudden deceleration judgment process, which is usually set to several seconds. Within the second time window, the sudden deceleration state of the vehicle is continuously checked to determine whether it occurs frequently; the counter is used to record the number of sudden decelerations of the vehicle within the second time window. Each time a sudden deceleration is detected, the value of the counter increases by 1; the ACCy value represents the acceleration of the vehicle in the Y-axis direction, reflecting the change in the acceleration of the vehicle in this direction. The ACCy value is used to detect the sudden deceleration state of the vehicle.

[0064] After the vehicle decelerates suddenly and enters the frequent sudden deceleration judgment process, a second time window is set. The length of the second time window is usually a few seconds, for example, it is set to 5 seconds, 10 seconds or longer, depending on the specific application; at the same time, a counter is used to record the number of sudden decelerations detected in the second time window. The initial value of the counter is set to 0. Each time a sudden deceleration is detected, the counter will increase by 1.

[0065] By setting a second time window and continuously monitoring the acceleration data (ACCy value) of the vehicle within the time window, it is detected whether the vehicle has frequent sudden deceleration. When a sudden deceleration of the target vehicle is detected, the value of the counter is increased by 1. If the vehicle has multiple sudden decelerations within the second time window (that is, the counter value exceeds the set threshold), it means that the vehicle has frequent sudden decelerations. Further decision-making mechanisms (such as issuing warnings to the driver, recording data, etc.) are used to process it, which helps to enhance driving safety and remind the driver to improve his driving habits.

[0066] Furthermore, the present application method also includes:

[0067] Based on the second time window and the counter, it is determined whether a preset frequent sudden deceleration constraint is satisfied; if satisfied, it is determined that the target vehicle is in a frequent sudden deceleration state.

[0068] Specifically, the second time window refers to the time interval used to monitor whether the vehicle has multiple sudden decelerations during the frequent sudden deceleration judgment process. This time window is set to a number of seconds, usually between a few seconds and tens of seconds. The second time window is used to record the number of sudden decelerations of the vehicle. Each time a sudden deceleration is detected, the counter will accumulate by 1. The function of the counter is to help determine whether the vehicle has frequent sudden decelerations within the set time.

[0069] The preset frequent sudden deceleration constraint refers to the pre-set conditions or standards in the frequent sudden deceleration judgment, which are usually based on the number of sudden decelerations. For example, within a fixed time window, the number of sudden decelerations of the vehicle must exceed a certain threshold to be judged as frequent sudden deceleration. The frequent sudden deceleration state refers to the situation where the vehicle frequently decelerates suddenly within a certain time range, which may indicate that the driver has bad driving habits or may cause safety hazards. The presence of frequent sudden deceleration behavior is determined by monitoring the vehicle's acceleration data.

[0070] In the second time window, the sudden deceleration state of the vehicle is continuously monitored, and the number of sudden decelerations is recorded using a counter. After the monitoring of the second time window is completed, it is checked whether the value of the counter has reached the preset frequent sudden deceleration constraint; if the value of the counter has reached the preset sudden deceleration number threshold, it is considered that the target vehicle is already in a frequent sudden deceleration state. At this time, the corresponding operation is performed, such as issuing a warning message to the driver, or recording the event for subsequent analysis.

[0071] By combining the second time window and the counter, the number of sudden decelerations of the vehicle is compared with the preset frequent sudden deceleration constraint; when the number of sudden decelerations exceeds the threshold, it is judged that the target vehicle has frequent sudden deceleration behavior, thereby helping to identify bad driving habits and improve driving safety. On this basis, warnings, records, feedback and other measures are taken to ensure that the driver can notice his improper driving behavior and improve safe driving.

[0072] Furthermore, the present application method also includes:

[0073] Based on the current rapid deceleration node, obtain the accumulated bit of the adjacent counter corresponding to the previous node; if the accumulated bit of the adjacent counter corresponding to the previous node is 0, set the accumulated bit of the counter of the current rapid deceleration node to 1, and record the timestamp corresponding to the current rapid deceleration node.

[0074] Specifically, a sudden deceleration node refers to a point in time when a vehicle suddenly decelerates during driving. Sudden deceleration usually refers to a sharp drop in the speed of the vehicle in a short period of time. It is a rapid deceleration caused by the driver suddenly braking or other reasons. The adjacent counter accumulation bit is a flag bit used to track the relationship between the front and rear sudden deceleration nodes. Each sudden deceleration node has an adjacent counter accumulation bit, indicating whether the node is associated with the previous sudden deceleration node. When the vehicle decelerates continuously, the current sudden deceleration node will be associated with the previous sudden deceleration node, marked as the adjacent counter accumulation bit is 1.

[0075] Each sudden deceleration node will have a counter, and the counter accumulation bit is used to record the number of sudden decelerations that occur at the node. When two sudden deceleration nodes occur consecutively, the counter accumulation bit of the current node will be set to 1, indicating that the current sudden deceleration is adjacent to the previous sudden deceleration node; the timestamp records the exact time when each sudden deceleration node occurs. The timestamp is used to determine the specific time when the node suddenly decelerates, which is convenient for subsequent time difference calculation and event analysis; when a sudden deceleration event is detected, the event is marked as the current sudden deceleration node, and the status of the previous sudden deceleration node will be checked. Further, the adjacent counter accumulation bit of the previous sudden deceleration node is obtained. The function of the adjacent counter accumulation bit is to determine whether the current sudden deceleration node has a continuous relationship with the previous node.

[0076] If the cumulative bit of the adjacent counter of the previous node is 0, the cumulative bit of the counter of the current rapid deceleration node is set to 1, indicating that this is a new rapid deceleration event. At the same time, the timestamp corresponding to the current rapid deceleration node is recorded to identify the specific time when the event occurred. By obtaining the cumulative bit of the adjacent counter of the previous rapid deceleration node, it is determined whether the current rapid deceleration node is adjacent to the previous node. If the previous node is independent of the current node, the cumulative bit of the counter of the current node is set to 1, indicating a new rapid deceleration event, and the timestamp is recorded for the event to distinguish between continuous rapid deceleration events and independent rapid deceleration events, so as to more accurately detect and analyze the rapid deceleration state.

[0077] Furthermore, the present application method also includes:

[0078] If the accumulated bit of the adjacent counter corresponding to the previous node is 1, the timestamp corresponding to the current rapid deceleration node is updated; at the same time, the time difference between the timestamp corresponding to the current rapid deceleration node and the timestamp corresponding to the previous node is calculated; it is determined whether the time difference is less than or equal to the preset time threshold under the preset frequent rapid deceleration constraint; if so, the frequent rapid deceleration flag is set to 1.

[0079] Specifically, the accumulated bit of the adjacent counter indicates whether the previous rapid deceleration node and the current rapid deceleration node are continuous events. If the accumulated bit of the adjacent counter of the previous node is 1, it means that the previous node and the current node are continuous rapid deceleration events. Otherwise, if it is 0, it means that the rapid deceleration event is an independent event. The timestamp is used to identify the moment when the current rapid deceleration node occurs, and is compared with the timestamp of the previous rapid deceleration node to calculate the time difference. The time difference refers to the difference between the occurrence times of two rapid deceleration events. Calculating the time difference can help the system determine the interval between the two rapid deceleration events, thereby analyzing whether it meets the restrictions on frequent rapid deceleration.

[0080] The preset frequent sudden deceleration constraint refers to a predetermined rule or condition used to determine whether sudden deceleration occurs frequently, for example, whether the number of sudden decelerations occurring within a certain period of time or the time interval between sudden deceleration events is less than a certain threshold; the frequent sudden deceleration flag is a flag used to indicate whether frequent sudden deceleration events are detected. If the conditions for frequent sudden deceleration are detected, the flag is set to 1, indicating that frequent sudden deceleration occurs.

[0081] When the current rapid deceleration node is detected, the adjacency counter accumulation bit of the previous rapid deceleration node is checked. If the adjacency counter accumulation bit of the previous node is 1, it indicates that the current rapid deceleration node and the previous node are continuous rapid deceleration events. In this case, the timestamp of the current node will be updated to the specific time when the current rapid deceleration event occurs. After updating the timestamp, the time difference between the timestamp of the current rapid deceleration node and the timestamp of the previous rapid deceleration node is calculated. The time difference between the timestamp corresponding to the current rapid deceleration node and the timestamp corresponding to the previous node represents the interval time between the occurrence of these two rapid deceleration events.

[0082] After calculating the time difference, it is judged according to the preset frequent sudden deceleration constraint. If the time difference is less than or equal to the preset time threshold under the preset frequent sudden deceleration constraint, it means that the interval between the current sudden deceleration event and the previous sudden deceleration event is very short, and the frequent sudden deceleration flag is set to 1. At this time, it has been determined that the target vehicle is in a frequent sudden deceleration state, marking the frequent sudden deceleration of the vehicle, providing more accurate driving analysis and safety warning.

[0083] In summary, the beneficial effects of the embodiments of the present application are:

[0084] This application deploys IMU sensors to synchronously collect vehicle acceleration and angular velocity data, and after multi-level filtering processing, uses the first time window to analyze the driving speed, determines the frequent sudden deceleration state, sends an early warning to the driver, and resets the state for the next monitoring. It achieves the technical effect of efficiently and synchronously collecting the vehicle's acceleration and angular velocity data through the IMU sensor, judging the sudden deceleration state according to the ACCy value, and judging the road surface flatness in combination with the ACCz value, accurately identifying the vehicle's sudden deceleration behavior, effectively reducing the misjudgment rate, and improving the accuracy of sudden deceleration detection.

[0085] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0086] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A vehicle driving speed dynamic monitoring method based on IMU, characterized in that: The method comprises: Based on the target vehicle, an IMU sensor is deployed, and a first sampling trigger signal corresponding to the IMU sensor and a second sampling trigger signal corresponding to the vehicle speed sensor are synchronous pulses; Based on a first sampling trigger signal corresponding to the IMU sensor and a second sampling trigger signal corresponding to the vehicle speed sensor, the acceleration data and angular velocity data of the target vehicle during driving are collected; Performing multi-level filtering on the acceleration data and the angular velocity data, and analyzing the driving speed data using a first time window; Determine whether the target vehicle is in a frequent rapid deceleration state through the driving speed data, and obtain a determination result; Based on the judgment results, a warning message is sent to the driver, and the status is reset to prepare for the next monitoring.

2. The method according to claim 1, characterized in that The IMU sensor obtains the raw data of the six-axis gyroscope through the SPI interface; Processing the raw data by low-pass filtering to obtain first-layer processed data; Based on the first layer processed data, Kalman filtering is performed to obtain the second layer processed data.

3. The method according to claim 2, characterized in that Acquiring raw data of the six-axis gyroscope, the method further comprises: Based on the six-axis gyroscope, axial fusion is performed to determine the three-axis Euler angle; Based on the three-axis Euler angles, two bidirectional linked lists are created, wherein the number of bits of each linked list is N, which are used to store ACCy floating point numbers and ACCz floating point numbers respectively.

4. The method according to claim 3, characterized in that Each node contains two floating point numbers and pointers to the previous and next nodes to support bidirectional traversal; The two floating point numbers include a pair of an ACCy floating point number and an ACCz floating point number.

5. The method according to claim 4, characterized in that The two floating point numbers include a pair of an ACCy floating point number and an ACCz floating point number, and the method includes: Based on the sampling frequency, configure the first time window; Based on the first time window, when performing data collection, the currently collected ACCy value and ACCz value are added to the data set; At the same time, a sliding window mechanism is adopted to remove the original ACCy value and the original ACCz value in the data set each time new data is added.

6. The method according to claim 5, characterized in that The method comprises: Calculating the average value of the ACCy values ​​in the data set to obtain an average ACCy value; Setting a sudden deceleration judgment threshold, and comparing the average ACCy value with the sudden deceleration judgment threshold; If the average ACCy value is less than the sudden deceleration judgment threshold, it is determined that the target vehicle is in a sudden deceleration state.

7. The method according to claim 6, characterized in that By using the driving speed data, determining whether the target vehicle is in a frequent rapid deceleration state, the method further includes: After determining that the target vehicle is in a rapid deceleration state, the frequent rapid deceleration judgment process is started as follows: Setting a second time window and counter corresponding to frequent rapid deceleration detection; In the second time window, the ACCy value is continuously monitored based on the driving speed data, and the counter is incremented by 1 each time a sudden deceleration state is detected.

8. The method according to claim 7, characterized in that The method comprises: Based on the second time window and the counter, determining whether a preset frequent sudden deceleration constraint is satisfied; If the conditions are met, it is determined that the target vehicle is in a frequent rapid deceleration state.

9. The method according to claim 8, characterized in that Continuously monitor the ACCy value, and increment the counter by 1 each time a sudden deceleration state is detected. The method includes: Based on the current rapid deceleration node, obtain the accumulated bit of the adjacent counter corresponding to the previous node; If the cumulative bit of the adjacent counter corresponding to the previous node is 0, the cumulative bit of the counter of the current rapid deceleration node is set to 1, and the timestamp corresponding to the current rapid deceleration node is recorded.

10. The method according to claim 9, characterized in that Determining whether a preset frequent rapid deceleration constraint is met, the method further includes: If the accumulated bit of the adjacent counter corresponding to the previous node is 1, then the timestamp corresponding to the current rapid deceleration node is updated; At the same time, the time difference between the timestamp corresponding to the current rapid deceleration node and the timestamp corresponding to the previous node is calculated; Determining whether the time difference is less than or equal to a preset time threshold under the preset frequent rapid deceleration constraint; If so, the frequent rapid deceleration flag is set to 1.

Images