Driving behavior safety monitoring system

Abstract

The invention discloses a driving behavior safety monitoring system, and belongs to the technical field of driving safety monitoring equipment. Which comprises a data processing module, a test module and a prediction module, and comprises the following steps: S1, the data processing module collects, processes and samples original driving behavior data; s2, performing simulation training on the data sampled in the step S1 by using a test module; and S3, comparing and fusing the real track of the driving behavior and the predicted value of the simulation training result in the step S2, and finally outputting the result, so that the judgment accuracy is high, and the system can continuously learn and follow up according to the historical driving behavior to obtain a more accurate test track and improve the judgment accuracy.

Description

Technical Field

[0001] The present invention provides a driving behavior safety monitoring system, which belongs to the technical field of driving safety monitoring equipment. Background Art

[0002] Coal mining and transportation vehicles have a significant impact on safety, and the driver's behavior directly affects the incidence of accidents. To ensure the safety of various vehicles in the mine, a safety monitoring system for driving behavior is extremely important.

[0003] Existing monitoring systems build a personnel positioning and safety behavior monitoring system, and implement 3D-GIS display and operation within this system. This system displays the precise location of personnel and vehicles, along with relevant information such as their direction of travel, current speed, and driver, in real time. This allows for data collection, storage, analysis, and fusion processing, presenting alarm information in a visual interface to notify relevant mine personnel. This improves the dispatching, command, and emergency response capabilities of the mine dispatch center, ensuring the safe production of all types of mine transport vehicles. However, existing monitoring systems can only monitor and analyze based on the precise location of personnel and vehicles, their direction of travel, current speed, and driver, and lack comparisons with previous driving behaviors. This inadequate learning ability can easily lead to misjudgments. To address these issues, a driving behavior safety monitoring system is needed. Summary of the Invention

[0004] In order to solve the above problems, the present invention proposes a technical solution: a driving behavior safety monitoring system, including a data processing module, a test module, and a prediction module, including the following steps: S1, the data processing module collects, processes, and samples the original driving behavior data; S2, the data sampled in step S1 is simulated and trained using the test module; S3, the actual trajectory of the driving behavior and the predicted value of the simulation training result in step S2 are compared and integrated, and the result is finally output.

[0005] As an improvement, the data processing module in step S1 includes an original data set module, a grid processing module, and an equidistant sampling module.

[0006] As an improvement, the original data set module includes a large amount of data on previous normal driving behavior. After collecting the previous driving behavior data, it is processed using data gridding, and finally the data is sampled at equal intervals using the equal-interval sampling module and then transmitted to the test module.

[0007] As an improvement, in step S2, a recurrent neural network is used to simulate and train the data sampled in step S1 to obtain a test trajectory.

[0008] As an improvement, the recurrent neural network is an LSTM (Long Short-Term Memory Network).

[0009] As an improvement, in step S3, the actual driving behavior trajectory is comprehensively compared with the test trajectory obtained in step S2, and a predicted value is finally obtained and output.

[0010] Beneficial effects of the present invention:

[0011] The data processing module collects, processes, and samples the original driving behavior data, and uses the neural network LSTM to simulate and train the sampled data. This can obtain the test trajectory under normal conditions and make a comprehensive comparison with the actual driving behavior trajectory. The judgment accuracy is high, and the system can continuously learn and follow up based on historical driving behavior to obtain more accurate test trajectories and improve judgment accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 The figure is a structural diagram of a driving behavior safety monitoring system of the present invention. DETAILED DESCRIPTION

[0014] The present invention will be further described below with reference to the accompanying drawings.

[0015] according to Figure 1 As shown: The present invention provides a driving behavior safety monitoring system: including a data processing module, a test module, and a prediction module, including the following steps: S1, the data processing module collects, processes, and samples the original driving behavior data; S2, the data sampled in step S1 is simulated and trained using the test module; S3, the actual trajectory of the driving behavior and the predicted value of the simulation training result in step S2 are compared and integrated, and the result is finally output.

[0016] The data processing module in step S1 includes a raw dataset module, a gridding processing module, and an evenly spaced sampling module. The raw dataset module contains a large amount of data on past normal driving behavior. After collecting this data, it is then processed using data gridding. Gridding can unify data from different sources and formats into a common grid structure, facilitating data integration and analysis. It is also suitable for efficient matrix operations, which can accelerate data processing, allowing for better management and analysis of spatial data and improving data processing efficiency and accuracy. Finally, the evenly spaced sampling module samples the data at even intervals and then transfers it to the testing module. Evenly spaced sampling ensures a more even distribution of the data after sampling, avoiding excessive or insufficient data in certain areas, thereby better reflecting the overall characteristics of the data.

[0017] In step S2, a recurrent neural network (LSTM) is used to simulate and train the data sampled in step S1 to generate a test trajectory. By using the LSTM (Long Short-Term Memory) network within a recurrent neural network for data processing, the complex patterns and long-term dependencies in time series data can be better captured, improving the efficiency and accuracy of data processing.

[0018] In step S3, the actual driving behavior trajectory is comprehensively compared with the test trajectory obtained in step S2, and a predicted value is finally obtained and output.

[0019] When the driving behavior safety monitoring system provided by the present application is used normally, first, various monitoring devices can monitor the driver's driving data in real time, including the precise location of personnel and vehicles, the direction of travel, the current speed, etc. During real-time monitoring, a data set of a certain range is collected from the original data set module. After being processed by the grid processing module, the data is sampled at equal intervals using the equidistant sampling module and then transmitted to the test module; the sampled data is simulated and trained using the recurrent neural network LSTM to obtain a test trajectory; the obtained test trajectory is then comprehensively compared with the current driving behavior trajectory, and finally a predicted value is obtained and output. During processing, the system can improve the accuracy of driving behavior prediction by continuously comparing and analyzing historical data and integrating it with real-time driving behavior.

[0020] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. A driving behavior safety monitoring system, characterized in that: It includes a data processing module, a test module, and a prediction module, and includes the following steps: S1, the data processing module collects, processes, and samples the original driving behavior data; S2, the test module performs simulation training on the data sampled in step S1; S3, the real trajectory of the driving behavior and the predicted value of the simulation training result in step S2 are compared and integrated, and the result is finally output.

2. A driving behavior safety monitoring system according to claim 1, characterized in that: The data processing module in step S1 includes an original data set module, a grid processing module, and an equidistant sampling module.

3. A driving behavior safety monitoring system according to claim 2, characterized in that: The original data set module includes a large amount of data on normal driving behavior in the past. After the data on the past driving behavior is collected, it is processed by data gridding, and finally the data is sampled at equal intervals by using an equal-interval sampling module, and then transmitted to the test module.

4. A driving behavior safety monitoring system according to claim 3, characterized in that: In step S2, a recurrent neural network is used to simulate and train the data sampled in step S1 to obtain a test trajectory.

5. A driving behavior safety monitoring system according to claim 4, characterized in that: The recurrent neural network is a LSTM (Long Short-Term Memory) network.

6. A driving behavior safety monitoring system according to claim 5, characterized in that: In step S3, the actual driving behavior trajectory is comprehensively compared with the test trajectory obtained in step S2, and finally a predicted value is obtained and output.

Images