An overcrowding monitoring system and method for driverless buses

Abstract

The present invention provides an overloading monitoring system and method for driverless buses. The system includes: a sensing module, a central data processing module, and an overloading prompt module; the sensing module is used to sense the passenger status and seat status to obtain status data; the central data processing module is used to fuse the status data based on the data contribution degree, and process the fused status data based on the overloading decision-making model to obtain an overloading decision; the overloading prompt module is used to perform an overloading prompt based on the overloading decision. The overloading monitoring system for driverless buses is an intelligent system integrating sensing, processing, and prompting. Its application will effectively improve the safety and operation efficiency of buses, and provide passengers with a more convenient and comfortable travel experience.

Description

Technical Field

[0001] The present invention belongs to the technical field of overloading monitoring of motor vehicles, and particularly relates to a system and method for overloading monitoring of driverless buses. Background Technique

[0002] In recent years, driverless vehicles have been applied in different regions across the country. The operation of passenger transportation without drivers inside the vehicle has become a representative mode of passenger transportation at present and in the future because it has significant advantages such as significantly reducing the employment cost of transportation operators and bringing a good riding experience to passengers. Considering that there are no traditional drivers or in-vehicle safety officers inside the driverless vehicle and there are no other staff members inside the vehicle, when the number of passengers exceeds the rated passenger capacity, it not only violates existing laws and regulations, but also affects driving safety and the normal operation of the vehicle.

[0003] The existing mainstream overloading monitoring methods are as follows: First, remote personnel monitor whether there is overloading by means of in-vehicle cameras, but there are problems such as being unable to accurately determine whether there is overloading due to insufficient light at night. Second, represented by the utility model CN220962591U, the number of people on the vehicle is obtained by installing information collection modules, communication modules, etc. on highways and determining whether there is overloading, but the cost of modifying road infrastructure is relatively high, the application scenario is limited to highways and not applicable to urban roads, and the recognition accuracy is easily affected by environmental factors such as driving weather. Third, represented by the utility model CN209785252, infrared thermal imagers and cameras are installed above the lane to obtain the number of people on the vehicle and determine whether there is overloading. Similarly, the cost is relatively high, and the recognition accuracy is easily affected by environmental factors such as driving weather. Fourth, represented by the utility model CN220962591U, the overloading situation is detected jointly by visible light cameras and single infrared cameras, but when there are errors in image recognition, it is impossible to determine whether there is overloading.

[0004] In summary, the above existing overloading monitoring methods have their respective technical drawbacks; second, they do not consider the characteristics of driverless buses at present, such as all passengers need to be seated and standing is prohibited; third, they do not consider the reasons for overloading of driverless buses, such as passengers temporarily deciding not to get off after arriving at the station and the number of passengers getting on the vehicle after the vehicle arrives at the station being more than the number of remaining seats. Therefore, the existing overloading monitoring schemes are not applicable to driverless buses without drivers inside. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the present invention provides a system and method for overloading monitoring of driverless buses, realizing accurate monitoring of the overloading situation of driverless vehicles and ensuring the safety of passengers.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] An overloading monitoring system for driverless buses, comprising: a sensing module, a central data processing module, and an overloading prompting module;

[0008] The sensing module is used to sense the passenger status and seat status and obtain status data;

[0009] The central data processing module is used to fuse the status data based on the data contribution degree, and process the fused status data based on an overloading decision-making model to obtain an overloading decision;

[0010] The overloading prompting module is used to perform overloading prompting based on the overloading decision.

[0011] Preferably, the sensing module includes a door infrared sensor unit, a passenger seating area detection unit, a non-passenger seating area detection unit, and a passenger area seat pressure detection unit;

[0012] The door infrared sensor unit is used to record the number of passengers getting on and off the bus;

[0013] The passenger seating area detection unit is used to detect the first occupancy situation of the seats and identify the category of the occupancy object;

[0014] The non-passenger seating area detection unit is used to detect the standing situation of passengers in the carriage;

[0015] The passenger area seat pressure detection unit is used to detect the second occupancy situation of the seats;

[0016] Among them, the first occupancy situation of the seats and the second occupancy situation of the seats are weighted to obtain the final occupancy situation of the seats.

[0017] Preferably, the passenger seating area detection unit includes:

[0018] An image acquisition sub-unit, which is used to acquire the seat image of the passenger seating area, perform enhancement processing on the seat image of the passenger seating area, and obtain an enhanced seat image;

[0019] A feature extraction sub-unit, which is used to construct a feature extraction network based on a deformable convolution block and a C2f module, and perform feature extraction on the enhanced seat image based on the feature extraction network to obtain a multi-scale seat feature image;

[0020] A feature enhancement sub-unit, which is used to fuse the multi-scale seat feature images based on a PANet feature enhancement network through a bottom-up path and a top-down path to obtain a fused feature image;

[0021] An anchor box allocation sub-unit, which is used to adjust the anchor box size based on preset requirements, and perform anchor box allocation for the fused feature image based on a PAA module;

[0022] The seat occupancy detection unit is used to establish a detection head by combining a spatial attention mechanism and a convolution operation, and predict the bounding box regression value of the assigned anchor box based on the detection head to obtain the first seat occupancy situation; wherein, the first seat occupancy situation includes occupied seats and empty seats.

[0023] The seat occupancy object classification unit is used to perform global average pooling on the fused feature image to obtain a feature vector of a preset length; classify the seat occupancy object based on the feature vector to obtain the occupied object category; wherein, the occupied object category includes passenger occupancy and item occupancy.

[0024] Preferably, the central data processing module includes a data receiving unit, a data fusion unit, and an overcrowding decision-making unit.

[0025] The data receiving unit is used to receive the status data, wherein the status data includes the number of passenger getting on and off, the final seat occupancy situation, the occupied object category, and the standing situation of passengers in the carriage.

[0026] The data analysis unit is used to analyze the number of passenger getting on and off, calculate the number of passengers getting on, the number of passengers getting off, and the current total number of passengers in the vehicle; analyze the final seat occupancy situation and the occupied object category to obtain the number of seated passengers and the remaining number of seats; analyze the standing situation of passengers in the carriage to obtain the number of standing passengers.

[0027] The data fusion unit is used to fuse the data analysis results to obtain fused data.

[0028] The overcrowding decision-making unit is used to make an overcrowding decision for the driverless bus based on the fused data and the overcrowding decision-making model.

[0029] Preferably, the data fusion unit includes:

[0030] Perform denoising processing and outlier detection on the data analysis results to obtain the processed data analysis results.

[0031] Unify the format of the processed data analysis results and correspond to the time stamp to obtain multi-dimensional time series data.

[0032] Extract features from the multi-dimensional time series data to obtain data features.

[0033] Calculate and sort the overcrowding decision-making contribution degrees of the data features to obtain the data features that meet the preset contribution degree threshold.

[0034] Based on a deep neural network, perform a new feature representation on the data features that meet the preset contribution degree threshold to obtain the fused data.

[0035] Preferably, the overloading decision-making unit includes:

[0036] A model construction unit, configured to construct the overloading decision-making model based on the type of driverless bus, the passenger capacity standard, and the existing status data set by using a deep learning algorithm;

[0037] A model calculation unit, configured to process the data to be fused for decision-making based on the overloading decision-making model to obtain the overloading probability and the degree of overloading;

[0038] A decision-making judgment unit, configured to judge the overloading probability and the degree of overloading based on the corresponding driverless bus type and the passenger capacity standard to obtain an overloading decision.

[0039] Preferably, the overloading prompt module includes an external passenger overloading prompt unit, a background overloading prompt unit, and a bus overloading prompt unit;

[0040] The external passenger overloading prompt unit is configured to give an overloading prompt to the passengers waiting to board the bus outside based on the overloading decision;

[0041] The background overloading prompt unit is configured to give an overloading prompt to the backstage staff of the driverless bus based on the overloading decision;

[0042] The bus overloading prompt unit is configured to give an overloading prompt to the vehicle central control platform of the driverless bus based on the overloading decision.

[0043] The present invention further provides a method for monitoring overloading of a driverless bus for implementing the method, including the following steps:

[0044] Perceive the passenger status and the seat status to obtain status data;

[0045] Fuse the status data based on the data contribution degree, and process the fused status data based on the overloading decision-making model to obtain an overloading decision;

[0046] Give an overloading prompt based on the overloading decision.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows: Through the collaborative work of the sensing module and the central data processing module, the system can monitor the passenger capacity of the bus in real time, and issue an early warning immediately when overloading is detected, effectively avoiding potential safety hazards caused by overloading. The system can accurately judge the passenger capacity of the bus, thereby improving the operation efficiency and service quality of the bus. Through the overloading reminder module, the system can timely inform passengers of the passenger capacity of the bus, helping passengers make reasonable travel choices and avoiding inconveniences and troubles caused by overloading. The data collected and analysis results of the system can provide strong support for the intelligent management of the bus company, contributing to the optimization of bus routes, the adjustment of departure intervals and other decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the present invention, the accompanying drawings required for use in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other accompanying drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a schematic structural diagram of an overloading monitoring system for driverless buses according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0051] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Embodiment 1

[0053] As Figure 1 shown, an overloading monitoring system for driverless buses includes: a sensing module, a central data processing module, and an overloading reminder module;

[0054] The sensing module is used to sense the passenger status and seat status to obtain status data. A further implementation is that the sensing module includes a door infrared sensor unit, a passenger seating area detection unit, a non-passenger seating area detection unit, and a passenger area seat pressure detection unit;

[0055] The door infrared sensor unit is used to record the number of passengers getting on and off the bus. The door infrared sensor is installed at the door.

[0056] A passenger seating area detection unit, which is used to detect the first occupancy situation of the seats and identify the category of the occupancy object; in this embodiment, the passenger seating area detection unit adopts a passenger seating area camera group, which is installed in the vehicle and can detect the position of each seat. A further implementation manner is that the passenger seating area detection unit includes:

[0057] An image acquisition sub-unit, which is used to acquire the seat images in the passenger seating area and perform enhancement processing on the seat images in the passenger seating area to obtain enhanced seat images; specifically, during the driving process of a driverless bus, there are daytime driving, night driving, and driving in a low-light environment. Therefore, in the case of a complex lighting environment, the present invention establishes a low-light image enhancement model to process the acquired low-light images and improve the image detection accuracy. The specific implementation steps are as follows:

[0058] Based on an improved VGG neural network model for intercepting the feature information of the seat images in the passenger seating area and a RetinexNet neural network model for enhancing the brightness of the image background, a low-light image enhancement model is constructed; among them, the VGG neural network model is improved by weight transfer. The improved VGG neural network model includes convolutional kernels, channels, and Haar wavelet transform layers; when using weight transfer to improve the VGG neural network model, different weights are set for the content loss and the style loss, and the image is migrated to the corresponding part of the model according to the weight values of different losses of the image for image optimization and enhancement. At the same time, multiple iterative trainings are performed to ensure the fusion balance of the image at the content level and the style level.

[0059] The RetinexNet neural network model includes a decomposition network and an enhancement network. The decomposition network adopts a five-layer convolutional neural network, receives a pair of image inputs (low illuminance and normal images), and outputs illumination and reflection images. Through this step, the model can effectively separate the illumination and reflection components in the image, laying a foundation for subsequent enhancement processing. The enhancement network is a deeper network, using a nine-layer convolutional structure to further enhance the brightness and quality of the decomposed reflection image. In addition, a noise reduction step is included in the network to ensure that no additional noise is introduced during the image enhancement process and to maintain the natural attributes of the image.

[0060] A feature extraction sub-unit, which is used to construct a feature extraction network based on deformable convolutional blocks and C2f modules, and perform feature extraction on the enhanced seat images based on the feature extraction network to obtain multi-scale seat feature images;

[0061] A feature enhancement sub-unit, which is used to fuse the multi-scale seat feature images based on the PANet feature enhancement network through a bottom-up path and a top-down path to obtain a fused feature image;

[0062] An anchor box allocation subunit, configured to adjust the anchor box size based on preset requirements, and perform anchor box allocation for the fused feature image based on the PAA module;

[0063] An occupancy detection unit, configured to establish a detection head by combining a spatial attention mechanism and a convolution operation, and predict the bounding box regression value of the allocated anchor box based on the detection head to obtain the first occupancy situation of the seat; wherein, the first occupancy situation of the seat includes occupied and empty seats;

[0064] An occupancy object classification unit, configured to perform global average pooling on the fused feature image using a global average pooling layer to obtain a feature vector of a preset length; classify the occupancy object based on the feature vector to obtain the occupancy object category; wherein, the occupancy object category includes passenger occupancy and item occupancy.

[0065] A non-passenger seating area detection unit, configured to detect the standing situation of passengers in the carriage; in this embodiment, the non-passenger seating area detection unit uses a non-passenger seating area camera group installed in the vehicle.

[0066] A passenger area seat pressure detection unit, configured to detect the second occupancy situation of the seat; in this embodiment, the passenger area seat pressure detection unit uses a passenger area seat pressure sensor installed under each seat.

[0067] Wherein, the first occupancy situation of the seat and the second occupancy situation of the seat are weighted using preset weights to obtain the final seat occupancy situation. In the present invention, in order to overcome the detection defects that may be caused by simple image acquisition in the prior art, a seat pressure sensor is added, and the real-time data collected by the pressure sensor is weighted with the image target detection result to obtain the final seat occupancy situation, improving the seat occupancy detection accuracy.

[0068] A central data processing module, configured to fuse status data based on data contribution degree, and process the fused status data based on an overcrowding decision model to obtain an overcrowding decision;

[0069] A further implementation manner is that the central data processing module includes a data receiving unit, a data fusion unit, and an overcrowding decision unit;

[0070] A data receiving unit, configured to receive status data, wherein the status data includes the number of passenger getting on and off, the final seat occupancy situation, the occupancy object category, and the standing situation of passengers in the carriage;

[0071] A data analysis unit, configured to analyze the number of passenger getting on and off, calculate the number of passengers getting on, the number of passengers getting off, and the current total number of passengers in the vehicle; analyze the final seat occupancy situation and the occupancy object category to obtain the number of seated passengers and the remaining number of seats; analyze the standing situation of passengers in the carriage to obtain the number of standing passengers;

[0072] A data fusion unit, which is used to fuse the data analysis results to obtain fused data;

[0073] An overcrowding decision-making unit, which is used to make an overcrowding decision for the driverless bus based on the fused data and the overcrowding decision-making model.

[0074] A further implementation manner is that the data fusion unit includes:

[0075] Perform denoising processing and outlier detection on the data analysis results to obtain the processed data analysis results; specifically, adopt smoothing processing to remove the noise generated in the data due to various reasons (such as measurement errors, transmission errors, etc.), and improve the accuracy and reliability of the data. Use the Isolation Forest algorithm to identify and process the outliers in the data (i.e., the values that are significantly different from most data points), and these outliers may be generated due to reasons such as incorrect records, data corruption, or special events.

[0076] Unify the format of the processed data analysis results and correspond to the time stamps to obtain multi-dimensional time series data; specifically, assign a unique time stamp to each data point to ensure that the data can be arranged and analyzed in chronological order.

[0077] Adopt Fourier transform to extract features from the multi-dimensional time series data to obtain data features;

[0078] Use the Gradient Boosting Tree to calculate and sort the overcrowding decision-making contribution degrees of the data features to obtain the data features that meet the preset contribution degree threshold; specifically, according to the contribution degree calculation results, sort the features from high to low, and screen out the features that meet the preset contribution degree threshold.

[0079] Based on the deep neural network, perform a new feature representation on the data features that meet the preset contribution degree threshold to obtain fused data. Specifically, use a convolutional neural network to encode and decode the selected features to learn a new feature representation. Concatenate and fuse the learned new feature representation with the original data features to obtain the concatenated data of the data analysis results of each category, and then concatenate the concatenated data of each category into a feature vector to obtain the final fused data.

[0080] A further implementation manner is that the overcrowding decision-making unit includes:

[0081] A model construction unit, which is used to construct an overloading decision model based on the type of driverless bus, the passenger capacity standard, and the existing status data set by using a deep learning algorithm; specifically, the convolutional layer of the convolutional neural network is used to extract local features from the fused data, the ReLU non-linear activation function is introduced in the activation layer, and the pooling layer downsamples the output of the convolutional layer to reduce the number of parameters and improve the calculation efficiency. The outputs of the convolutional layer and the pooling layer are converted into feature vectors of a preset length. A recurrent neural network is used to receive the feature vectors output by the convolutional neural network and time series data (such as the running time of the bus, the change in the number of passengers over time, etc.), and the LSTM is used to process the time series data to capture the dynamic characteristics and dependencies in the data. According to the preset task requirements, the predicted values of the overloading probability or the degree of overloading are output.

[0082] The outputs of the convolutional neural network and the recurrent neural network are weighted and fused based on preset weights, and an additional fully connected layer is introduced on the fused feature vectors to further extract and integrate features. The fully connected layer performs non-linear transformation on the fused features to extract higher-level feature representations. The weights and biases of the fully connected layer are optimized through the backpropagation algorithm during the training process.

[0083] A model calculation unit, which is used to process the to-be-decided fused data based on the overloading decision model to obtain the overloading probability and the degree of overloading;

[0084] A decision-making judgment unit, which is used to judge the overloading probability and the degree of overloading based on the corresponding driverless bus type and the passenger capacity standard to obtain an overloading decision. Specifically, the overloading probability and the degree of overloading output by the model calculation unit are compared with the set threshold. If the overloading probability exceeds the threshold or the degree of overloading reaches or exceeds the preset standard, it is judged as overloading. Specifically, for example, outside the seated area, as long as someone is standing, it is considered overloading; if there is no one standing and all the seats are occupied, it is judged as about to be overloaded.

[0085] An overloading prompt module, which is used to give an overloading prompt based on the overloading decision.

[0086] A further implementation manner is that the overloading prompt module includes an out-of-vehicle passenger overloading prompt unit, a background overloading prompt unit, and a bus overloading prompt unit;

[0087] The out-of-vehicle passenger overloading prompt unit is used to give an overloading prompt to the passengers waiting to board outside the vehicle based on the overloading decision; the overloading prompt for the passengers boarding outside the vehicle includes using methods such as sound, light, and electrical signals at the door to remind and warn the passengers who still want to board outside the vehicle when the vehicle is full.

[0088] The background overcrowding prompt unit is used to prompt the backstage staff of the driverless bus based on the overcrowding decision; it uses 5G transmission and other methods to remind relevant personnel in the enterprise backstage that the vehicle is overcrowded.

[0089] The bus overcrowding prompt unit is used to prompt the vehicle central control platform of the driverless bus based on the overcrowding decision. The overcrowding prompt for driverless vehicles includes transmitting the overcrowding information to the vehicle central control platform by means of CAN signals, wireless local area network transmission, etc., and transmitting control instructions such as suppressing vehicle startup and continuous door opening to the driverless system until the passengers in the vehicle return to normal.

[0090] The data collected by the system and the analysis results can provide strong support for the intelligent management of the bus company, and help optimize bus routes, adjust departure intervals and other decision-making.

[0091] On the other hand, the present invention also provides a method for monitoring overcrowding of a driverless bus, which is used to implement the method, including the following steps:

[0092] Perceive the passenger status and seat status to obtain status data;

[0093] Based on the data contribution degree, fuse the status data, and process the fused status data based on the overcrowding decision-making model to obtain an overcrowding decision;

[0094] Based on the overcrowding decision, give an overcrowding prompt.

[0095] The embodiments described above are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. An overcrowding monitoring system for driverless buses, characterized in that, Including: A perception module, a central data processing module, and an overcrowding prompt module; The perception module is used to perceive the passenger status and seat status and obtain status data; The central data processing module is used to fuse the status data based on the data contribution degree, and process the fused status data based on the overcrowding decision-making model to obtain an overcrowding decision; The overcrowding prompt module is used to perform an overcrowding prompt based on the overcrowding decision; The perception module includes a door infrared sensor unit, a passenger seating area detection unit, a non-passenger seating area detection unit, and a passenger area seat pressure detection unit; The door infrared sensor unit is used to record the number of passengers getting on and off the vehicle; The passenger seating area detection unit is used to detect the first occupancy situation of the seat and identify the category of the occupancy object; The non-passenger seating area detection unit is used to detect the standing situation of passengers in the carriage; The passenger area seat pressure detection unit is used to detect the second occupancy situation of the seat; Wherein, the first occupancy situation of the seat and the second occupancy situation of the seat are weighted to obtain the final seat occupancy situation; The central data processing module includes a data receiving unit, a data fusion unit, and an overcrowding decision unit; The data receiving unit is used to receive the status data, wherein the status data includes the number of passengers getting on and off the vehicle, the final seat occupancy situation, the category of the occupancy object, and the standing situation of passengers in the carriage; The data analysis unit is used to analyze the number of passengers getting on and off the vehicle, calculate the number of passengers getting on, the number of passengers getting off, and the current total number of passengers in the vehicle; analyze the final seat occupancy situation and the category of the occupancy object to obtain the number of seated passengers and the remaining number of seats; analyze the standing situation of passengers in the carriage to obtain the number of standing passengers; The data fusion unit is used to fuse the data analysis results to obtain fusion data; The overcrowding decision unit is used to make an overcrowding decision on the driverless bus based on the fusion data and the overcrowding decision-making model; The data fusion unit includes: Perform denoising processing and outlier detection on the data analysis results to obtain processed data analysis results; Unify the format of the processed data analysis results and correspond to the time stamp to obtain multi-dimensional time series data; Extract features from the multi-dimensional time series data to obtain data features; Calculate and sort the overcrowding decision contribution degrees of the data features to obtain data features that meet the preset contribution degree threshold; Based on a deep neural network, perform a new feature representation on the data features that meet the preset contribution degree threshold to obtain the fusion data.

2. The overcrowding monitoring system for driverless buses according to claim 1, wherein, The passenger seating area detection unit includes: An image acquisition sub-unit, which is used to acquire the seat image of the passenger seating area, and perform enhancement processing on the seat image of the passenger seating area to obtain an enhanced seat image; A feature extraction sub-unit, which is used to construct a feature extraction network based on a deformable convolution block and a C2f module, and perform feature extraction on the enhanced seat image based on the feature extraction network to obtain a multi-scale seat feature image; A feature enhancer unit, which is used to fuse multi-scale seat feature images based on the PANet feature enhancement network through a bottom-up path and a top-down path to obtain a fused feature image; An anchor box assignment unit, which is used to adjust the anchor box size based on preset requirements and assign anchor boxes to the fused feature image based on the PAA module; An occupancy detection unit, which is used to establish a detection head by combining a spatial attention mechanism and a convolution operation, and predict the bounding box regression value of the assigned anchor box based on the detection head to obtain the first occupancy situation of the seat; wherein, the first occupancy situation of the seat includes occupied and empty seats; An occupancy object classification unit, which is used to perform global average pooling on the fused feature image by using a global average pooling layer to obtain a feature vector of a preset length; classify the occupancy object based on the feature vector to obtain the occupancy object category; wherein, the occupancy object category includes passenger occupancy and item occupancy.

3. The overloading monitoring system for driverless buses according to claim 1, characterized in that, The overcrowding decision-making unit includes: A model construction unit, which is used to construct the overcrowding decision-making model by using a deep learning algorithm based on the driverless bus type, the passenger capacity standard, and the existing state data set; A model calculation unit, which is used to process the to-be-decided fusion data based on the overcrowding decision-making model to obtain the overcrowding probability and the overcrowding degree; A decision-making judgment unit, which is used to judge the overcrowding probability and the overcrowding degree based on the corresponding driverless bus type and the passenger capacity standard to obtain an overcrowding decision.

4. The overloading monitoring system for driverless buses according to claim 1, wherein, The overcrowding prompt module includes an out-of-vehicle passenger overcrowding prompt unit, a background overcrowding prompt unit, and a bus overcrowding prompt unit; The out-of-vehicle passenger overcrowding prompt unit is used to prompt the passengers waiting to board the vehicle outside the vehicle based on the overcrowding decision; The background overcrowding prompt unit is used to prompt the backstage staff of the driverless bus based on the overcrowding decision; The bus overcrowding prompt unit is used to prompt the vehicle central control platform of the driverless bus based on the overcrowding decision.

5. A method for monitoring overloading of driverless buses, which is used to implement the system described in any one of claims 1-4, characterized in that, It includes the following steps: Perceive the passenger state and the seat state to obtain state data; Fuse the state data based on the data contribution degree, and process the fused state data based on the overcrowding decision-making model to obtain an overcrowding decision; Perform an overcrowding prompt based on the overcrowding decision.

Images