Hand-held intelligent terminal vehicle scheduling system and method integrated with Internet of Vehicles

Abstract

The invention discloses an Internet of Vehicles integrated handheld intelligent terminal vehicle scheduling system and method, and relates to the technical field of vehicle scheduling, and the system comprises a sensor deployment module which obtains the position, speed, acceleration and driver operation behaviors of a vehicle in real time through a GPS, an inertial sensor and a vehicle-mounted camera; the factor risk coupling module is used for constructing a traffic fluctuation depth model and predicting a multi-factor coupling risk; the adaptive scheduling controller design module is used for dynamically adjusting path planning through global path conflicts and congestion risks; and the dynamic task allocation module adjusts vehicle task allocation in real time according to the output of the controller. According to the method, the defects of an existing vehicle networking scheduling system in the aspects of information acquisition, risk prediction, response speed and resource allocation are overcome, so that the accuracy, the real-time performance and the system robustness of vehicle scheduling are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle dispatching, and specifically provides a vehicle dispatching system and method for a handheld intelligent terminal integrated with the Internet of Vehicles. Background Art

[0002] The Internet of Vehicles system relies on a variety of sensors and wireless communication technologies, such as GPS, inertial sensors, and in-vehicle cameras, etc., to achieve real-time monitoring of vehicle status; however, due to factors such as high-speed vehicle movement, communication network congestion, and bad weather, the stability and real-time nature of data transmission are often difficult to guarantee, thereby affecting the timely collection and feedback of vehicle dynamic information by the system.

[0003] In the existing dispatching system during the task assignment and path planning processes, it mostly relies on preset rules and static parameters, and has insufficient response capabilities to real-time traffic conditions and emergencies; traditional dispatching strategies often lack an adaptive mechanism and cannot be dynamically adjusted according to changes in the actual traffic environment, resulting in a significant reduction in dispatching effects and resource utilization efficiency during peak hours or under bad weather conditions.

[0004] In the actual traffic environment, there are complex mutual influence relationships among drivers, vehicles, and road conditions; existing systems usually have difficulty comprehensively considering the coupling effects among the three, especially when dealing with the synergistic effects of driver operation behaviors and road risk factors, information islands and decision-making biases are likely to occur, thus affecting overall traffic safety and dispatching efficiency.

[0005] Therefore, the present invention provides a vehicle dispatching system and method for a handheld intelligent terminal integrated with the Internet of Vehicles. Summary of the Invention

[0006] The purpose of the present invention is to provide a vehicle dispatching system and method for a handheld intelligent terminal integrated with the Internet of Vehicles to solve the existing problems raised in the above background art.

[0007] To achieve the above purpose, the present invention provides the following technical solution: A vehicle dispatching system for a handheld intelligent terminal integrated with the Internet of Vehicles, including the following steps:

[0008] A sensor deployment module that obtains vehicle position, speed, acceleration, and driver operation behaviors in real time through GPS, inertial sensors, and in-vehicle cameras;

[0009] A factor risk coupling module for constructing a traffic fluctuation depth model to predict multi-factor coupling risks;

[0010] An adaptive dispatching controller design module that dynamically adjusts path planning through global path conflicts and congestion risks;

[0011] The dynamic task allocation module adjusts vehicle task allocation in real time according to the output of the controller.

[0012] A further improvement of the present invention lies in that the sensor deployment module includes a terminal integration unit, a roadside unit, and a data fusion unit. The terminal integration unit includes an in-vehicle terminal and a handheld terminal. The in-vehicle terminal integrates a GPS, an inertial sensor, and an in-vehicle camera to collect vehicle position, speed, acceleration, and driver operation behavior in real time. The handheld terminal is used to receive driver preferences and task urgency. The roadside unit is used to deploy microwave radars, cameras, and meteorological sensors to monitor road congestion indices, weather conditions, and accident information. The data fusion unit is used to construct a multi-dimensional input sequence.

[0013] A further improvement of the present invention lies in that the factor risk coupling module includes a special time node marking unit and a traffic fluctuation depth model construction unit. The special time node marking unit is used to insert a marker tag during peak hours and accident-prone hours h , and the marker is converted into a high-dimensional vector through linear projection and input into the model together with the original data.

[0014] A further improvement of the present invention lies in that the traffic fluctuation depth model construction unit includes a multi-head attention encoder that linearly projects the multi-dimensional input sequence to a unified dimension to form an input sequence Input, captures the dynamic associations among vehicles, traffic, environment, and drivers through the multi-head attention mechanism, and further extracts features through a feed-forward neural network. The model outputs the traffic risk probability distribution for the next n steps: where y T+n represents the predicted traffic congestion index at the nth moment, W n (z) represents a dynamic mapping matrix that adjusts the weight according to the input z, and z represents the weather condition.

[0015] A further improvement of the present invention lies in that the adaptive scheduling controller design module includes a fast response layer and a dynamic optimization layer. A traffic risk threshold is set. When is greater than or equal to the traffic risk threshold, the fast response layer is triggered. When is less than the traffic risk threshold, the dynamic optimization layer is triggered.

[0016] A further improvement of the present invention lies in that the fast response layer aims to minimize global path conflicts and time deviations, sets the objective function with constraints such as the path length being less than the vehicle's endurance limit, the vehicle speed being less than the rated speed threshold, and emergency tasks being preferentially allocated, and outputs candidate paths and control instructions.

[0017] A further improvement of the present invention lies in that the dynamic optimization layer includes randomly generating a group of particles, each particle representing a candidate path. The weight of each path is determined by a fitness function, and the calculation formula of the fitness function is:

[0018]

[0019] Among them, T safe (φ) represents the safety score of the φ-th path, which is determined by the accident probability of the road section. T effic (φ) represents the efficiency score of the φ-th path, which is normalized by the average vehicle speed; subsequently, the cumulative weight array is initialized, and the weights are accumulated by traversing the particle set; a random number is generated, and high-fitness particles are selected according to the cumulative weight; the selected particles are copied to a new set, and low-weight particles are eliminated, and the candidate path and control instructions are output.

[0020] A further improvement of the present invention lies in that, according to a vehicle scheduling system of a handheld intelligent terminal integrated with a vehicle network described in claim 1, it is characterized in that: the dynamic task allocation module includes allocating high-priority material transportation tasks to vehicles on low-risk paths and dynamically adjusting in combination with real-time road conditions; through a game theory model, idle vehicles are guided to high-demand areas.

[0021] On the other hand, the present invention provides a vehicle scheduling method for a handheld intelligent terminal integrated with a vehicle network, and the specific steps include:

[0022] S1. Real-time obtain the vehicle position, speed, acceleration, and driver operation behavior through GPS, inertial sensors, and in-vehicle cameras;

[0023] S2. Used to construct a traffic fluctuation depth model to predict multi-factor coupling risks;

[0024] S3. Dynamically adjust the path planning through the global path conflict and congestion risks;

[0025] S4. Real-time adjust the vehicle task allocation according to the output of the controller.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. The present invention first constructs a traffic fluctuation depth model through the factor risk coupling module, and combines the multi-head attention encoder to capture the dynamic associations among vehicles, traffic, environment, and drivers, improving the system's prediction ability for multi-factor coupling risks. Especially, the sensitivity of the model to peak hours, bad weather, and high-accident periods is enhanced through the special time node marking unit, enabling the system to more accurately predict the future traffic congestion index and providing data support for scheduling optimization;

[0028] 2. An adaptive scheduling controller design module is adopted, which combines a fast response layer and a dynamic optimization layer to achieve intelligent scheduling for different traffic environments. In the case of low traffic risk, the system adopts a coarse scheduling layer with low computational complexity to quickly generate a scheduling plan and improve the system response speed; in a complex traffic environment, the system triggers the fine scheduling layer, which provides more refined scheduling optimization by optimizing computing resources and improves the scheduling accuracy.

[0029] 3. Through a path planning optimization algorithm, dynamic adjustment is performed based on global path conflicts and congestion risks. The fast response layer adjusts the driving strategy in real time by minimizing the objective function of global path conflicts and time deviation, considering factors such as path length, vehicle speed, and congestion score; the dynamic optimization layer optimizes the path through a particle screening mechanism to ensure that the selected path is optimal in terms of safety and traffic efficiency, reducing the vehicle scheduling cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a framework diagram of a vehicle scheduling system for a vehicle - connected integrated handheld intelligent terminal according to the present invention;

[0031] Figure 2 It is a flowchart of a vehicle scheduling method for a vehicle - connected integrated handheld intelligent terminal according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The technical solution of the present invention will be described in detail below with reference to the drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. Without conflict, the technical features in the embodiments of the present invention and the embodiments can be combined with each other.

[0033] The term "and / or" merely describes an association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects before and after.

[0034] Embodiment 1

[0035] Figure 1 A framework diagram of a vehicle scheduling system for a vehicle - connected integrated handheld intelligent terminal disclosed in this embodiment is shown, which is characterized by including the following steps:

[0036] A sensor deployment module that obtains the vehicle position, speed, acceleration, and driver operation behavior in real time through GPS, inertial sensors, and in - vehicle cameras;

[0037] A factor risk coupling module for constructing a traffic fluctuation depth model to predict multi - factor coupling risks;

[0038] Adaptive Scheduling Controller Design Module, which dynamically adjusts path planning based on global path conflicts and congestion risks;

[0039] Dynamic Task Allocation Module, which adjusts vehicle task allocation in real time according to the controller output.

[0040] The sensor deployment module includes a terminal integration unit, a roadside unit, and a data fusion unit. The terminal integration unit includes an in-vehicle terminal and a handheld terminal. The in-vehicle terminal integrates GPS, an inertial sensor, and an in-vehicle camera to collect vehicle position, speed, acceleration, and driver operation behavior in real time. The handheld terminal is used to receive driver preferences and task urgency; the roadside unit is used to deploy microwave radars, cameras, and meteorological sensors to monitor road congestion indexes, weather conditions, and accident information; the data fusion unit is used to construct a multi-dimensional input sequence X = {X P , X T , X E , X D}, where X P represents the vehicle dynamic data set, including position, speed, and acceleration; X T represents the traffic flow and road condition data set, including congestion index and accident point coordinates; X E represents the environmental data, including temperature, wind speed, and light intensity; X D represents the driver behavior and task priority data, representing the number of hard brakes and the task deadline.

[0041] The factor risk coupling module includes a special time node marking unit and a traffic fluctuation depth model construction unit; the special time node marking unit is used to insert a marker tag h during peak hours and accident-prone hours (rainy days and nights) to enhance the model's sensitivity to emergencies. The marker is converted into a high-dimensional vector through linear projection and input into the model together with the original data;

[0042] The traffic fluctuation depth model construction unit includes a multi-head attention encoder, which linearly projects the multi-dimensional input sequence to a unified dimension to form an input sequence Input, captures the dynamic associations among vehicles, traffic, environment, and drivers through the multi-head attention mechanism, and further extracts features through a feed-forward neural network. The model outputs the traffic risk probability distribution for the next n steps: Among them, y T+n represents the predicted traffic congestion index at the nth moment, and W n(z) represents the dynamic mapping matrix, which adjusts the weights according to the input z. z represents the weather conditions. Discrete conditions (such as weather types) are converted into high-dimensional vectors using one-hot encoding or an embedding layer. For continuous conditions (such as temperature and wind speed), normalization or piecewise encoding is used.

[0043] The adaptive scheduling controller design module includes a fast response layer and a dynamic optimization layer; a traffic risk threshold is set. When is greater than or equal to the traffic risk threshold, the fast response layer is triggered. When is less than the traffic risk threshold, the dynamic optimization layer is triggered.

[0044] The preliminary plan is quickly generated through the coarse scheduling layer to ensure the system response speed; the fine scheduling layer provides fine optimization in complex environments to ensure the scheduling accuracy.

[0045] In a regular environment, the system only runs the coarse scheduling layer to reduce the consumption of computing resources; in a complex environment, the system triggers the fine scheduling layer to concentrate resources to solve difficult problems.

[0046] The hierarchical design enables the system to adapt to various traffic scenarios; the dynamic optimization ability of the fine scheduling layer improves the robustness of the system.

[0047] The fast response layer aims to minimize the global path conflict and time deviation, and sets the objective function with the path length less than the vehicle's endurance limit, the vehicle speed less than the rated speed threshold, and the priority allocation of emergency tasks as constraints, and outputs the candidate path and control instructions. The objective function is expressed as: where represents the deviation between the actual and predicted arrival times, represents the current path length, represents the ideal path length, ψ con (φ) represents the congestion score of section φ, which is calculated from the real-time traffic flow. α1, α2, and α3 are weight coefficients to balance time, path efficiency, and congestion cost.

[0048] The control instructions include suggesting that the vehicle accelerate or decelerate to adapt to the path planning, indicating that the vehicle turns or goes straight at the next intersection, and dynamically adjusting the driving strategy of the vehicle according to the urgency of the task.

[0049] The dynamic optimization layer includes randomly generating a group of particles, each particle representing a candidate path. The weight of each path is determined by the fitness function. The calculation formula of the fitness function is:

[0050]

[0051] where T safe(φ) represents the safety score of the φ-th path, which is determined by the accident probability of the road section, T effic (φ) represents the efficiency score of the φ-th path, which is obtained by normalizing the average vehicle speed; subsequently, initialize the cumulative weight array, traverse the particle set to accumulate weights; generate a random number, and select high-fitness particles according to the cumulative weights; copy the selected particles to a new set, eliminate low-weight particles, and output the candidate path and control instructions.

[0052] The calculation formula for the path safety score is Among them, P acc represents the accident probability of road section i, which is obtained by fitting historical data with real-time weather, d i represents the length of road section i, δ i is the weight of road section i. For example, the weights of tunnel and bridge sections are higher.

[0053] The dynamic task allocation module includes allocating high-priority material transportation tasks to vehicles on low-risk paths and dynamically adjusting them in combination with real-time road conditions; through the game theory model, guiding idle vehicles to high-demand areas such as business districts and transportation hubs to maximize resource utilization.

[0054] The setting of the threshold and weight can be based on the default settings of the present invention or can be set by the operator himself.

[0055] Embodiment 2

[0056] Figure 2 shows a flowchart of a vehicle scheduling method for a vehicle networking integrated handheld intelligent terminal according to the present invention. Based on the same inventive concept as Embodiment 1, the present invention provides a vehicle scheduling method for a vehicle networking integrated handheld intelligent terminal. The specific steps include:

[0057] S1. Real-time obtain the vehicle position, speed, acceleration, and driver operation behavior through GPS, inertial sensors, and in-vehicle cameras;

[0058] S2. Used to construct a traffic fluctuation depth model to predict multi-factor coupling risks;

[0059] S3. Dynamically adjust the path planning through the global path conflict and congestion risks;

[0060] S4. According to the output of the controller, adjust the vehicle task allocation in real time.

[0061] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0062] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0063] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0064] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0065] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims. These all fall within the protection scope of the present invention.

Claims

1. A vehicle dispatching system with a handheld intelligent terminal integrated with an Internet of Vehicles, characterized in that: The following steps are involved: The sensor deployment module uses GPS, inertial sensors, and on-board cameras to obtain vehicle position, speed, acceleration, and driver operation behavior in real time; Factor risk coupling module, used to build a traffic fluctuation depth model and predict multi-factor coupling risks; Adaptive scheduling controller design module, dynamically adjusts path planning based on global path conflicts and congestion risks; The dynamic task allocation module adjusts the vehicle task allocation in real time according to the controller output.

2. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 1 is characterized by: The sensor deployment module includes a terminal integration unit, a roadside unit and a data fusion unit. The terminal integration unit includes a vehicle-mounted terminal and a handheld terminal. The vehicle-mounted terminal integrates GPS, inertial sensors and vehicle-mounted cameras to collect vehicle position, speed, acceleration and driver operation behavior in real time. The handheld terminal is used to receive driver preferences and task urgency. The roadside unit is used to deploy microwave radars, cameras and meteorological sensors to monitor road congestion index, weather conditions and accident information; the data fusion unit is used to construct a multi-dimensional input sequence.

3. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 1 is characterized in that: The factor risk coupling module includes a special time node marking unit and a traffic fluctuation depth model construction unit; the special time node marking unit is used to insert a tag during peak hours and accident-prone periods. h , the labels are converted into high-dimensional vectors through linear projection and input into the model together with the original data.

4. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 3 is characterized by: The traffic fluctuation deep model construction unit includes a multi-head attention encoder, which linearly projects the multi-dimensional input sequence to a unified dimension to form an input sequence Input. The dynamic relationship between vehicles, traffic, environment and drivers is captured through the multi-head attention mechanism, and features are further extracted through a feedforward neural network. The model outputs the probability distribution of traffic risks in the next n steps: Among them, y T+n represents the traffic congestion index predicted at the nth moment, W n (z) represents the dynamic mapping matrix, and the weights are adjusted according to the input z, where z represents the weather conditions.

5. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 1 is characterized by: The adaptive dispatch controller design module includes a fast response layer and a dynamic optimization layer; setting a traffic risk threshold When it is greater than or equal to the traffic risk threshold, the rapid response layer is triggered. When it is less than the traffic risk threshold, the dynamic optimization layer is triggered.

6. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 5 is characterized by: The rapid response layer aims to minimize global path conflicts and time deviations, sets the objective function with path length less than the vehicle endurance limit, vehicle speed less than the rated speed threshold, and priority allocation of emergency tasks as constraints, and outputs candidate paths and control instructions.

7. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 5 is characterized by: The dynamic optimization layer includes randomly generating a group of particles, each particle represents a candidate path, and the weight of each path is determined by the fitness function. The fitness function calculation formula is: Among them, T safe (φ) represents the safety score of the φth path, which is determined by the accident probability of the road section, T effic (φ) represents the efficiency score of the φth path, which is obtained by normalizing the average vehicle speed. Then, the cumulative weight array is initialized, and the particle set is traversed to accumulate the weights. Random numbers are generated, and high-fitness particles are selected according to the cumulative weights. The selected particles are copied to the new set, low-weight particles are eliminated, and candidate paths and control instructions are output.

8. The vehicle dispatching system of a handheld intelligent terminal integrated with a vehicle network according to claim 1 is characterized by: The dynamic task allocation module includes allocating high-priority material transportation tasks to vehicles on low-risk paths and dynamically adjusting them based on real-time road conditions; and guiding idle vehicles to high-demand areas through a game theory model.

9. A vehicle dispatching method for a handheld intelligent terminal integrated with a vehicle networking, used to execute a vehicle dispatching system for a handheld intelligent terminal integrated with a vehicle networking as claimed in any one of claims 1 to 8, characterized in that: The specific steps include: S1, obtain vehicle position, speed, acceleration, and driver operation behavior in real time through GPS, inertial sensors, and on-board cameras; S2, used to build a traffic fluctuation depth model and predict multi-factor coupling risks; S3, dynamically adjust path planning based on global path conflicts and congestion risks; S4. Adjust vehicle task allocation in real time according to controller output.

Images