An Efficient Intelligent Obstacle Avoidance Method for Unmanned Systems Based on a Heterogeneous Brain Model

Abstract

The present application discloses an efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model, as well as a training method, device, storage medium, equipment, and computer program product of the heterogeneous brain model, including: generating input feature data for inputting into the heterogeneous brain model according to driving sampling data collected from sensors of the unmanned system; inputting the input feature data into the heterogeneous brain model to initialize each neuron in the heterogeneous brain model; running the heterogeneous brain model, and determining the target output state data of the target neuron in the run heterogeneous brain model as the model output data of the heterogeneous brain model; controlling the navigation state of the unmanned system according to the model output data. The present application makes the operation process of data interpretable by using the corresponding relationship between neurons in the heterogeneous brain model, realizes real-time obstacle avoidance on a low-power small embedded device, provides an intuitive explanation of the decision-making process, and improves the safety and reliability of the unmanned system.

Description

Technical Field

[0001] This application belongs to the field of driverless, and particularly relates to an efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model, as well as a training method, device, storage medium, equipment, and computer program product of the heterogeneous brain model. Background Art

[0002] The rapid development of driverless technology has led to the transformation of intelligent transportation systems, and unmanned systems are increasingly widely used in various environments. In scenarios such as intelligent transportation, automated production, and drone cruising, the efficient obstacle avoidance ability of unmanned systems is crucial for improving their autonomy and safety. Therefore, how to achieve efficient and intelligent obstacle avoidance in complex and changing environments has become an important research topic in the field of unmanned systems.

[0003] Currently, among the obstacle avoidance methods for unmanned systems, there is an obstacle avoidance method based on a deep convolutional neural network, which utilizes the powerful feature extraction ability of the deep convolutional neural network to automatically learn obstacle features from the perception image and generate obstacle avoidance decisions.

[0004] However, the method based on a deep convolutional neural network requires a large amount of data and computing resources during training and inference, and it is difficult to achieve real-time obstacle avoidance on small embedded devices. At the same time, due to the "black box" attribute of the deep convolutional neural network model, the decision-making process is difficult to intuitively explain, resulting in poor interpretability, which will affect the safety and reliability of unmanned systems. Summary of the Invention

[0005] This application aims to provide an efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model, as well as a training method, device, storage medium, equipment, and computer program product of the heterogeneous brain model, at least solving the problem of insufficient safety and reliability in the obstacle avoidance process of unmanned systems.

[0006] In a first aspect, an embodiment of this application discloses an efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model, which is applied to a controller of the unmanned system and includes:

[0007] Obtain driving sampling data collected from sensors of the unmanned system, and generate input feature data for inputting into the heterogeneous brain model according to the driving sampling data; the heterogeneous brain model includes multiple neural units; each input neural unit in the heterogeneous brain model has a corresponding output neural unit; each input neural unit is respectively used to obtain the state value of the output neural unit corresponding to the input neural unit, and solve the state value of the input neural unit according to the obtained state value;

[0008] Input the input feature data into the heterogeneous brain model to initialize each neural unit in the heterogeneous brain model;

[0009] Run the heterogeneous brain model, and determine the target output state data of the target neural unit in the heterogeneous brain model after running as the model output data of the heterogeneous brain model;

[0010] Control the navigation state of the unmanned system according to the model output data.

[0011] In a second aspect, an embodiment of the present application also discloses a training method for a heterogeneous brain model for efficient intelligent obstacle avoidance, which is used to train the heterogeneous brain model in the method for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model as described in the first aspect, including:

[0012] Divide the heterogeneous brain model into multiple data processing areas, so that each neural unit in the heterogeneous brain model has a corresponding data processing area, and establish a connection relationship between the neural units of every two data processing areas according to the initialized connection strength; the connection strength is used to characterize the connection quantity and connection weight between the multiple neural units included in the two data processing areas;

[0013] Train the connection strength to obtain a trained heterogeneous brain model.

[0014] In a third aspect, an embodiment of the present application also discloses a device for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model, which is applied to the controller of the unmanned system and includes:

[0015] An acquisition module, configured to acquire driving sampling data collected from sensors of the unmanned system, and generate input feature data for inputting into the heterogeneous brain model according to the driving sampling data; the heterogeneous brain model includes multiple neural units; each input neural unit in the heterogeneous brain model has a corresponding output neural unit; each input neural unit is respectively used to acquire the state value of the output neural unit corresponding to the input neural unit, and solve the state value of the input neural unit according to the acquired state value;

[0016] An input module, configured to input the input feature data into the heterogeneous brain model to initialize each neural unit in the heterogeneous brain model;

[0017] An output module, configured to run the heterogeneous brain model, and determine the target output state data of the target neural unit in the heterogeneous brain model after running as the model output data of the heterogeneous brain model;

[0018] A control module, configured to control the navigation state of the unmanned system according to the model output data.

[0019] Fourthly, an embodiment of the present application further discloses a training device for a heterogeneous brain model for efficient and intelligent obstacle avoidance, which is used to train the heterogeneous brain model in the method for efficient and intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model as described in the first aspect, including:

[0020] An initialization module, configured to divide the heterogeneous brain model into multiple data processing areas, so that each neuron in the heterogeneous brain model has a corresponding data processing area, and establish a connection relationship between the neurons of every two data processing areas according to the initialized connection strength; the connection strength is used to represent the connection quantity and connection weight between the multiple neurons included in two data processing areas;

[0021] A training module, configured to train the connection strength to obtain a trained heterogeneous brain model.

[0022] Fifthly, an embodiment of the present application further discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps described in the first aspect or the second aspect are implemented.

[0023] Sixthly, an embodiment of the present application further discloses an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor, and when the computer program is executed by the processor, the steps described in the first aspect or the second aspect are implemented.

[0024] Seventhly, an embodiment of the present application further discloses a computer program product, on which a computer program is stored, and when the computer program is executed by a processor, the steps described in the first aspect or the second aspect are implemented.

[0025] In summary, in the embodiments of the present application, by using the input feature data generated from obtaining driving sampling data and generating input feature data as the initial input of the heterogeneous brain model, the model can make full use of environmental perception data for efficient and intelligent obstacle avoidance. Furthermore, by inputting the input feature data into each neuron unit in the heterogeneous brain model, the initialization of the model is completed, enabling each neuron unit of the model to be in a ready state, so that the initialized neuron units can accurately reflect the characteristics of the input data. Then, by running the heterogeneous brain model, the operation process of the data can be interpreted by using the corresponding relationship between the neuron units in the heterogeneous brain model. While achieving real-time obstacle avoidance on low-power small embedded devices, it also overcomes the high computational complexity and "black box" property of deep convolutional neural networks, provides an intuitive explanation of the decision-making process while achieving efficient computing under low-power conditions. Furthermore, by determining the output state data of the target neuron unit, the navigation state of the unmanned system is controlled, realizing intelligent obstacle avoidance decision-making for the unmanned system and ensuring efficient obstacle avoidance in a complex and changing environment. Thus, based on the method of the embodiments of the present application, efficient and intelligent obstacle avoidance is achieved, improving the safety and reliability of the unmanned system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of illustrating the preferred embodiments and are not considered to be a limitation of the present application. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0027] Figure 1 is a flowchart of the steps of a method for efficient and intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model provided by an embodiment of the present application;

[0028] Figure 2 is a flowchart of the steps of another method for efficient and intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model provided by an embodiment of the present application;

[0029] Figure 3 is a determination process of an input feature data under an embodiment of the present application;

[0030] Figure 4 is an architecture of a heterogeneous brain model under an embodiment of the present application;

[0031] Figure 5 is a schematic diagram of a control mode of an unmanned system under an embodiment of the present application;

[0032] Figure 6 is a flowchart of the steps of a training method of a heterogeneous brain model for efficient and intelligent obstacle avoidance provided by an embodiment of the present application;

[0033] Figure 7 is the training configuration of a heterogeneous brain model provided by an embodiment of the present application;

[0034] Figure 8 is the flowchart of the steps of another training method of a heterogeneous brain model for efficient intelligent obstacle avoidance provided by an embodiment of the present application;

[0035] Figure 9 is the structural schematic diagram of an unmanned system efficient intelligent obstacle avoidance device based on a heterogeneous brain model provided by an embodiment of the present application;

[0036] Figure 10 is the structural schematic diagram of a training device of a heterogeneous brain model for efficient intelligent obstacle avoidance provided by an embodiment of the present application;

[0037] Figure 11 is the block diagram of an electronic device provided by an embodiment of the present application;

[0038] Figure 12 is the block diagram of another electronic device provided by an embodiment of the present application. Detailed implementation manners

[0039] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be fully conveyed to those skilled in the art.

[0040] Figure 1 is a method for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model provided by this embodiment, specifically including the following steps:

[0041] Step 101, obtain the driving sampling data collected from the sensors of the unmanned system, and generate input feature data for inputting into the heterogeneous brain model according to the driving sampling data.

[0042] Among them, the heterogeneous brain model includes multiple neural units; each input neural unit in the heterogeneous brain model has a corresponding output neural unit; each input neural unit is respectively used to obtain the state value of the output neural unit corresponding to the input neural unit, and solve the state value of the input neural unit according to the obtained state value.

[0043] In some embodiments of the present application, driving sampling data collected from sensors of an unmanned system is obtained, and input feature data for inputting into a heterogeneous brain model is generated based on the driving sampling data. This process includes: collecting environmental data obtained by the sensors of the unmanned system and converting this data into input feature data that can be used by the heterogeneous brain model. The input feature data refers to the specific data converted from the driving sampling data and used to initialize and drive the heterogeneous brain model. After performing this step, the input feature data provides a basis for the initialization and operation of subsequent models, enabling the model to make full use of environmental perception data for efficient and intelligent obstacle avoidance.

[0044] In a specific example, a driverless vehicle is equipped with Light Detection and Ranging (LiDAR) and camera sensors to obtain real-time environmental data. The system collects this data and converts it into an input feature data matrix through data processing algorithms. These input feature data include the position information and image features of obstacles in the environment. After performing the above execution process, the obtained input feature data can be directly used for the initialization of the heterogeneous brain model, enabling the model to accurately perceive the environment and make obstacle avoidance decisions.

[0045] Step 102: Input the input feature data into the heterogeneous brain model to initialize each neuron unit in the heterogeneous brain model.

[0046] In some embodiments of the present application, each neuron unit in the heterogeneous brain model is initialized by inputting the input feature data into the heterogeneous brain model. This process includes: passing the already generated input feature data to each neuron unit of the heterogeneous brain model. After performing this step, each neuron unit in the heterogeneous brain model is in a ready state, enabling the initialized neuron unit to accurately reflect the characteristics of the input data and providing a basis for subsequent model operation and decision-making.

[0047] In a specific example, the system converts the environmental data obtained by the sensors of the driverless vehicle into an input feature data matrix and inputs this data into each neuron unit of the heterogeneous brain model. Each neuron unit in the heterogeneous brain model is initialized according to the input data and is ready to receive and process subsequent environmental perception data. After performing the above execution process, the obtained initialized neuron unit accurately reflects the characteristics of the input data, enabling the heterogeneous brain model to efficiently make subsequent intelligent obstacle avoidance decisions.

[0048] Step 103: Run the heterogeneous brain model and determine the target output state data of the target neuron unit in the run heterogeneous brain model as the model output data of the heterogeneous brain model.

[0049] In some embodiments of the present application, by running a heterogeneous brain model and determining the target output state data of the target neural unit in the run heterogeneous brain model as the model output data of the heterogeneous brain model. This process includes: starting the heterogeneous brain model to simulate the operations and interactions between neurons until the target neural unit generates output state data. The target neural unit refers to the neuron in the heterogeneous brain model used to generate the final decision output. After performing this step, the model output data of the heterogeneous brain model will be used as the intelligent obstacle avoidance decision basis for the unmanned system to ensure efficient obstacle avoidance in a complex and changing environment.

[0050] In a specific example, the system has started the heterogeneous brain model and input the initialization data generated in the previous step. The neural units in the model generate the output state data of the target neural unit through a series of operations and interactions. The output state data of the target neural unit is determined as the model output data of the heterogeneous brain model and is used to guide the obstacle avoidance operation of the driverless vehicle. After performing according to the above execution process, the obtained model output data provides the information basis for the unmanned system to make intelligent obstacle avoidance decisions in the actual environment.

[0051] Step 104, controlling the navigation state of the unmanned system according to the model output data.

[0052] In some embodiments of the present application, by controlling the navigation state of the unmanned system according to the model output data. This process includes: receiving the model output data generated by the heterogeneous brain model and using these data to adjust and control the navigation state of the unmanned system. The model output data refers to the state data output by the target neural unit in the heterogeneous brain model. After performing this step, the unmanned system can perform intelligent obstacle avoidance according to the real-time environment and model decisions, thereby achieving efficient and safe navigation.

[0053] In a specific example, the driverless vehicle receives the model output data generated by the heterogeneous brain model, and this data contains the obstacle avoidance decision information for obstacles. The system then controls the direction and speed of the driverless vehicle according to these output data so that it can safely avoid the obstacles on the road. After performing according to the above execution process, the obtained adjustment and control information ensures that the driverless vehicle can achieve intelligent obstacle avoidance in a complex traffic environment.

[0054] In summary, in the embodiments of the present application, by using the input feature data generated by acquiring driving sampling data and generating input feature data as the initial input of the heterogeneous brain model, the model can make full use of environmental perception data for efficient and intelligent obstacle avoidance. Furthermore, by inputting the input feature data into each neuron in the heterogeneous brain model, the initialization of the model is completed, enabling each neuron in the model to be in a ready state, so that the initialized neurons can accurately reflect the characteristics of the input data. Then, by running the heterogeneous brain model, the operation process of the data can be made interpretable using the corresponding relationships between the neurons in the heterogeneous brain model. While achieving real-time obstacle avoidance on low-power small embedded devices, it also overcomes the high computational complexity and "black box" property of deep convolutional neural networks, providing an intuitive explanation of the decision-making process while achieving efficient computing under low-power conditions. Furthermore, by determining the output state data of the target neuron, the navigation state of the unmanned system is controlled, realizing intelligent obstacle avoidance decision-making for the unmanned system and ensuring efficient obstacle avoidance in complex and changing environments. Thus, based on the method of the embodiments of the present application, efficient and intelligent obstacle avoidance is achieved, improving the safety and reliability of the unmanned system.

[0055] Figure 2 This is another method for efficient and intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model provided in this embodiment, which specifically includes the following steps:

[0056] Step 201: Acquire driving sampling data collected from the sensors of the unmanned system, and generate input feature data for inputting into the heterogeneous brain model according to the driving sampling data.

[0057] Among them, the heterogeneous brain model includes multiple neurons; each input neuron in the heterogeneous brain model has a corresponding output neuron; each input neuron is used to obtain the state value of the output neuron corresponding to the input neuron, and solve the state value of the input neuron according to the obtained state value.

[0058] The method shown in this step has been described in step 101 and will not be elaborated here.

[0059] Optionally, as Figure 3 shown, the sensor is an image sensor, the driving sampling data is image driving sampling data A, and step 201 includes the following sub-steps:

[0060] Sub-step 2011: Downsample the image driving sampling data A to obtain multiple image block data B of the image driving sampling data.

[0061] In some embodiments of the present application, downsampling is performed on the image driving sampling data A to obtain a plurality of image block data B of the image driving sampling data. This process includes: reducing the resolution of the image or the number of pixels in the image from the original image data acquired by the sensor, and dividing it into a plurality of small image block data. Downsampling refers to the process of simplifying image data to improve the efficiency of subsequent processing by reducing unnecessary redundant information. After performing this step, the generated plurality of image block data B significantly reduces the data volume, thereby reducing the complexity of subsequent calculations and storage requirements.

[0062] In a specific example, an image in a video of an urban road scene that changes with time t is collected by an image sensor of a driverless vehicle as the image driving sampling data A. This video data is input into the system, and the system performs downsampling processing on each frame of the image, dividing the high-resolution image into several small image block data B. The resolution of each image block data B is 256×144. According to this execution process, a plurality of image block data B can finally be obtained, and these data blocks can be used for subsequent feature extraction and analysis.

[0063] Sub-step 2012, generate a data vector C for each image block data B, and determine the data matrix P obtained by concatenating the generated plurality of data vectors C as the input feature data.

[0064] In some embodiments of the present application, by generating a data vector C for each image block data B, and determining the data matrix P obtained by concatenating the generated plurality of data vectors C as the input feature data. This process includes: converting each image block data B into a corresponding data vector C, and these data vectors C are linear representation forms of the image block data. A data vector refers to a multi-dimensional vector transformed from image block data. Then, all the generated data vectors C are concatenated in sequence to form a data matrix P as the input feature data. After performing this step, the data matrix P contains the data features of all image blocks, providing complete environmental perception information for the subsequent model input.

[0065] In a specific example, the image collected by the image sensor of the driverless vehicle is downsampled into a plurality of image block data B with a resolution of 256×144. Each image block data B is expanded into a data vector C with a length of 64. Then, the data vectors C of 32×18 image blocks are concatenated in sequence to form a 576×64 data matrix P. After performing according to the above execution process, the obtained data matrix P is used as the input feature data, which contains the data features of all image blocks and can be used for the initialization of the heterogeneous brain model and intelligent obstacle avoidance decision-making.

[0066] Step 202: Input the input feature data into the heterogeneous brain model to initialize each neuron in the heterogeneous brain model.

[0067] The method shown in this step has been described in step 102 and will not be elaborated here.

[0068] Optionally, a corresponding second-order non-linear differential calculation model is preset in each neuron of this application. The second-order non-linear differential calculation model is characterized by the following formula set:

[0069] ;

[0070] ;

[0071] where, represents the state value of the i-th neuron; represents the state value of the j-th neuron; represents the recovery variable of the i-th neuron; represents the input feature data; represents the parameter of the linear growth rate of the neuron; represents the parameter of the non-linear effect of the heterogeneous brain model; represents the set of neurons connected to the i-th neuron; represents the transmission information weight between the j-th neuron and the i-th neuron; represents the perception weight of the input feature data to the i-th neuron; is the coupling parameter, indicating the influence degree on represents the decay parameter, controlling the recovery rate of represents the bias term of the heterogeneous brain model.

[0072] Through the preset second-order non-linear differential calculation model, this model can effectively simulate the response behavior of neurons in a complex environment, enhancing the adaptability and accuracy of the heterogeneous brain model. The reason for this design is to overcome the problems of large computational complexity and poor interpretability of traditional deep convolutional neural networks, enabling the model to achieve real-time obstacle avoidance under low-power conditions and providing an intuitive explanation of the decision-making process.

[0073] As Figure 4 shown, each black dot (Ni, i = 1, 2..., 20) in the figure represents a neuron. Further optionally, multiple neurons can be divided into six data processing areas: the input data acquisition processing area M1, the primary input data processing area M2, the advanced input data processing area M3, the integrated data processing area M4, the decision data processing area M5, and the action data processing area M6.

[0074] By reasonably dividing multiple neural units into different data processing areas, the above architecture can effectively organize and manage the data transmission and processing between neural units, improving the efficiency and accuracy of data processing. It can better achieve hierarchical processing and integration of data, optimize the overall performance of the neural network, and improve the intelligent obstacle avoidance ability of the system. Through this architecture, the system can achieve real-time obstacle avoidance under low power consumption conditions and provide an intuitive explanation of the decision-making process, overcoming the problems of large computational complexity and poor interpretability of traditional deep convolutional neural networks.

[0075] As a preferred embodiment of the present application, as Figure 4 shown, in a further embodiment of the present application, a heterogeneous brain model architecture is determined through the training process, which includes an acquisition input data processing area M1 containing 20 neural units (N1-N8), a primary input data processing area M2 containing 4 neural units (N17-N20), a high-level input data processing area M3 containing 3 neural units (N14-N16), an integrated data processing area M4 containing 2 neural units (N9, N10), a decision data processing area (M5) containing 2 neural units (N11, N12), and an action data processing area (M6) containing 1 neural unit (N13).

[0076] By reasonably dividing and training each data processing area, the above architecture can more efficiently manage and transmit data between neural units, thereby dividing a limited number of neural units, improving the processing efficiency and accuracy of the heterogeneous brain model. Through functional division and regional management of neural units, it can better achieve hierarchical processing and integration of data, optimize the overall performance of the neural network, and improve the intelligent obstacle avoidance ability of the unmanned system.

[0077] Optionally, as Figure 4 shown, based on the above second-order nonlinear differential calculation model, it can be set that when and only when the i-th neural unit belongs to the acquisition input data processing area, , to ensure that the acquisition input data processing area is used as the only external data (i.e., input feature data P) input module.

[0078] In the preferred embodiment of the present application, in order to clearly distinguish the functions of data processing areas, avoid confusion and redundancy in data input, thereby optimizing the performance of the neural network and enhancing the intelligent obstacle avoidance ability of the heterogeneous brain model, by ensuring that only the neural units in the acquisition input data processing area can receive external data, it can effectively control the source and processing flow of data input, improving the data processing efficiency and accuracy of the model. This ensures that the acquisition input data processing area is the only external data input module, making only the neural units in the acquisition input data processing area have non-zero perception weights.

[0079] Optionally, based on the above second-order non-linear differential calculation model, when solving the state value of the input neuron according to the obtained state value, the neuron completes the calculation of the state value through the following iterative steps:

[0080] Sub-step 2020, perform iteration on the following formula group according to the preset number of iterations to solve the state value of the input neuron:

[0081] ;

[0082] ;

[0083] ;

[0084] ;

[0085] where r and s represent intermediate variables of the algorithm, represents the step size of iterative solution, represents the moment corresponding to each state value; represents at the moment the state value of the input neuron, represents at the recovery variable of the input neuron at the moment.

[0086] In some embodiments of the present application, the state value of the input neuron is solved by performing iteration according to the preset number of iterations. This process includes: within the preset number of iterations, repeatedly perform the calculations in the formula group to gradually update the state value and the recovery variable of the input neuron. Iteration refers to a method of gradually approaching the final result by repeatedly executing calculation steps. After executing this step, the state value and the recovery variable of the input neuron are effectively updated, ensuring that the model can accurately reflect the characteristics of the input data. By presetting the number of iterations and the calculations of the formula group, the state value of the input neuron can be effectively updated and solved, thereby improving the calculation efficiency and accuracy of the model. The reason for this design is that by transforming the second-order non-linear differential model into an iterative model that can be quickly calculated through formulas, the dynamic behavior of neurons can be simulated more quickly, improving the real-time performance of the heterogeneous brain model in complex environments.

[0087] Step 203, run the heterogeneous brain model, and determine the target output state data of the target neuron in the run heterogeneous brain model as the model output data of the heterogeneous brain model.

[0088] The method shown in this step has been described in step 103 and will not be elaborated here.

[0089] Step 204, obtain the value of the target data component in the model output data.

[0090] In some embodiments of the present application, the value of the target data component in the model output data is obtained. This process includes: extracting the output component value of a specific target neuron from the model output data generated by the heterogeneous brain model. The target data component refers to the specific value in the model output data for controlling the navigation of the unmanned system. After performing this step, the obtained target data component will be used for the navigation control of the unmanned system in the subsequent steps to ensure that the system can make accurate obstacle avoidance decisions based on real-time environmental data.

[0091] In a specific example, after the system runs the heterogeneous brain model, the output component value of the target neuron in the model output data is extracted. This value represents the specific decision-making information during the obstacle avoidance process of the driverless vehicle, such as the steering angle or speed adjustment. After performing the above execution process, the obtained target data component value will be used in the control system of the driverless vehicle to ensure that it can perform safe and effective obstacle avoidance operations according to the real-time environment.

[0092] Step 205, control the navigation direction of the unmanned system according to the positive or negative situation of the value of the target data component, and control the navigation speed of the unmanned system according to the absolute value of the value of the target data component.

[0093] In some embodiments of the present application, the navigation direction of the unmanned system is controlled according to the positive or negative situation of the value of the target data component, and the navigation speed of the unmanned system is controlled according to the absolute value of the value of the target data component. This process includes: analyzing the value of the target data component to determine its positive or negative situation to decide in which direction the unmanned system should move. The target data component refers to the specific numerical value extracted from the model output data. Then, the navigation speed of the unmanned system is adjusted according to the absolute value of the target data component. After performing this step, the unmanned system can quickly adjust its navigation direction and speed according to environmental changes, achieving more flexible and efficient obstacle avoidance.

[0094] In a specific example, as Figure 5 shown, the control system of the unmanned aerial vehicle receives the output data of the model changing with time t. Assume that at a certain moment, the target data component is -0.8. The system determines that the unmanned aerial vehicle needs to move to the left according to the positive or negative situation of this value, and adjusts the turning speed of the unmanned aerial vehicle according to its absolute value (0.8). For example, a value of -0.8 may indicate that the unmanned aerial vehicle needs to move to the left at a speed of 0.8 meters per second. After performing the above execution process, the obtained adjustment information ensures that the unmanned aerial vehicle can quickly avoid obstacles on the road and achieve safe and reliable autonomous flight.

[0095] In summary, in the embodiment of the present application, by using the input feature data generated by acquiring driving sampling data and generating input feature data as the initial input of the heterogeneous brain model, the model can make full use of environmental perception data for efficient and intelligent obstacle avoidance. Furthermore, by inputting the input feature data into each neuron unit in the heterogeneous brain model, the initialization of the model is completed, so that each neuron unit of the model is in a ready state, and the initialized neuron unit can accurately reflect the characteristics of the input data. Then, by running the heterogeneous brain model, the operation process of the data can be made interpretable by using the corresponding relationship between the neuron units in the heterogeneous brain model. While achieving real-time obstacle avoidance on a low-power small embedded device, it also overcomes the high computational complexity and "black box" property of the deep convolutional neural network, realizes efficient computing under low-power conditions, and provides an intuitive explanation of the decision-making process. Furthermore, by determining the output state data of the target neuron unit, the navigation state of the unmanned system is controlled, and an intelligent obstacle avoidance decision for the unmanned system is realized, ensuring efficient obstacle avoidance in a complex and changing environment. Therefore, based on the method of the embodiment of the present application, efficient and intelligent obstacle avoidance is realized, and the safety and reliability of the unmanned system are improved.

[0096] Figure 6 This is a training method for a heterogeneous brain model for efficient and intelligent obstacle avoidance provided in this embodiment, which is used to train the heterogeneous brain model in the above-mentioned embodiment of the method for efficient and intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model, and specifically includes the following steps:

[0097] Step 301: Divide the heterogeneous brain model into multiple data processing areas, so that each neuron unit in the heterogeneous brain model has a corresponding data processing area, and establish a connection relationship between the neuron units of every two data processing areas according to the initialized connection strength.

[0098] Among them, the connection strength is used to represent the connection quantity and connection weight between multiple neuron units included in two data processing areas.

[0099] In some embodiments of the present application, the heterogeneous brain model is divided into multiple data processing regions, so that each neuron in the heterogeneous brain model has a corresponding data processing region, and the connection relationship between the neurons in every two data processing regions is established according to the initialized connection strength. This process includes: First, the entire heterogeneous brain model is divided into multiple data processing regions, so that each neuron in the model can correspond to a data processing region. Then, according to the preset connection strength, the connection relationship is established between the neurons in every two data processing regions. The connection strength is a parameter used to characterize the connection quantity and connection weight between multiple neurons included in two data processing regions. After performing this step, through reasonable setting of the connection strength, the connection of the neurons in the heterogeneous brain model becomes more reasonable and effective, laying a foundation for subsequent model training and operation.

[0100] In a specific example, as Figure 4 shown, the heterogeneous brain model of the unmanned system can be divided into six data processing regions: the acquisition input data processing region M1, the primary input data processing region M2, the high-level input data processing region M3, the integrated data processing region M4, the decision data processing region M5, and the action data processing region M6. Each data processing region contains a specific number of neurons. Then, according to the initialized connection strength, the connection relationship can be established between the neurons in each data processing region. The strength of these connection relationships characterizes the connection quantity and weight between each data processing region. After performing according to the above execution process, the obtained reasonable connection strength and connection relationship enable the heterogeneous brain model to have a good basic structure to support subsequent model training and operation for efficient intelligent obstacle avoidance.

[0101] Step 302: Train the connection strength to obtain a trained heterogeneous brain model.

[0102] In some embodiments of the present application, a trained heterogeneous brain model is obtained by training the connection strength. This process includes: According to a predetermined training algorithm, the connection strength between the data processing regions in the heterogeneous brain model is adjusted and optimized. The connection strength refers to the connection quantity and connection weight between multiple neurons included in two data processing regions. By continuously adjusting these connection parameters, the model can learn and optimize the information transmission path between neurons. After performing this step, the obtained trained heterogeneous brain model can process input data more efficiently and achieve intelligent obstacle avoidance.

[0103] In a specific example, the Gradient Descent Algorithm (GDA) can be used to train the connection strength of the heterogeneous brain model. During the training process, the model continuously adjusts the connection weights to minimize the prediction error. Monitor the performance of the model and optimize the connection strength according to the feedback information. After multiple iterative trainings, the trained heterogeneous brain model can accurately and quickly process the input data to achieve intelligent obstacle avoidance. After executing according to the above execution process, the trained heterogeneous brain model exhibits excellent obstacle avoidance performance in a complex and changeable environment.

[0104] As Figure 7 shown, it is the connection quantity and connection weight between each neuron in the heterogeneous brain model obtained by training a heterogeneous brain model (including 20 neurons) provided in an embodiment of the present application. In the figure, each square in the left square table represents the data transmission relationship (i.e., the connection quantity relationship) between any two neurons; and the depth value of the square determines the connection weight between the two neurons; the upper half of the table is used to represent the activation influence (positive weight) of the transmitted data on the neuron, and the lower half of the table is used to represent the inhibition influence (negative weight) of the transmitted data on the neuron.

[0105] In summary, in the embodiment of the present application, by obtaining the untranslated original text of the streaming text and inputting it into the semantic evaluation model to obtain the semantic integrity score of the target single sentence, it is possible to determine whether the target single sentence is a complete semantic unit, thereby deciding whether to update the untranslated original text; when the target single sentence meets the preset conditions, the untranslated original text is input into the machine translation model to obtain the translation result of the streaming text. Thus, based on the method of the embodiment of the present application, when performing machine translation of the streaming text, it is not necessary to wait for the entire text to be input to complete, and real-time translation of the streaming text can be achieved, solving the problems in the related art that due to the uncertain input time of the streaming text, it is difficult to perform effective segmentation and translation, resulting in low translation efficiency and poor translation quality. In addition, in the embodiment of the present application, by using the semantic evaluation model to evaluate the semantic integrity of the target single sentence, translation errors or ambiguities caused by incorrect single sentence segmentation can be effectively avoided.

[0106] Figure 8 This is another training method for a heterogeneous brain model for efficient intelligent obstacle avoidance provided in this embodiment, which is used to train the heterogeneous brain model in the above embodiment of the method for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model, and specifically includes the following steps:

[0107] Step 401, divide the heterogeneous brain model into multiple data processing areas according to the number of neurons corresponding to each data processing area.

[0108] Among them, a corresponding connection quantity and probability model are preset between each data processing area and the target data processing area.

[0109] In some embodiments of the present application, the heterogeneous brain model is divided into multiple data processing areas according to the number of neural units corresponding to each data processing area. This process includes: dividing the heterogeneous brain model into several data processing areas according to the number of neural units in the heterogeneous brain model, so that each data processing area contains a specific number of neural units. A corresponding connection quantity and probability model are preset between each data processing area and the target data processing area. These probability models characterize the connection probability and connection quantity between data processing areas. After performing this step, the obtained multiple data processing areas can manage and connect neural units more effectively, which helps subsequent training and model optimization.

[0110] In a specific example, the heterogeneous brain model of the unmanned system is divided into multiple data processing areas. And according to the number of neural units and functional requirements of each data processing area, the connection quantity and probability model between each area are set. For example, the connection quantity and probability model from the visual data processing area to the primary visual data processing area may be set such that each visual neuron is connected to 2 primary visual neurons, and the connection probability is 50%. After performing according to the above execution process, the connection relationship between the obtained multiple data processing areas is reasonable, improving the processing efficiency and accuracy of the model.

[0111] Step 402, regard each neural unit in the data processing area as an output neural unit, and determine multiple target neural units in the target data processing area as the input neural units of the output neural unit according to the probability model and connection quantity between the data processing area and the target data processing area, and establish a data connection between the output neural unit and the input neural unit.

[0112] In some embodiments of the present application, by regarding each neural unit in the data processing area as an output neural unit, multiple target neural units are determined in the target data processing area as the input neural units of the output neural unit according to the probability model and connection quantity between the data processing area and the target data processing area, and a data connection between the output neural unit and the input neural unit is established. This process includes: first, mark each neural unit in the data processing area as an output neural unit. Then, according to the set probability model and connection quantity, select multiple target neural units in the target data processing area as the corresponding input neural units. Data connection refers to the information transmission path established between the output neural unit and the input neural unit. After performing this step, the connection relationship between the neural units in the data processing area is clear, supporting the model to effectively transmit and process data in subsequent steps.

[0113] In a specific example, some neural units in the heterogeneous brain model are marked as output neural units. According to a preset probability model and the number of connections, multiple target neural units are selected as the input neural units for these output neural units. For example, each visual neural unit is connected to 2 primary visual neural units, and the connection probability is 50%. Then, by establishing data connections between these output neural units and input neural units, it is ensured that information can be effectively transmitted between each neural unit.

[0114] Step 403: According to the probability model corresponding to the data connection between the corresponding output neural unit and the input neural unit, establish the activation weight between the output data of the corresponding output neural unit and the input data of the input neural unit.

[0115] In some embodiments of the present application, by according to the probability model corresponding to the data connection between the corresponding output neural unit and the input neural unit, the activation weight between the output data of the corresponding output neural unit and the input data of the input neural unit is established. This process includes: analyzing the preset probability model and the number of connections, and allocating the activation weight between the output neural unit and the input neural unit according to this information. The activation weight is a parameter used to characterize the connection strength and information transmission efficiency in the neural unit connection. After performing this step, the activation weight between the output data and the input data is established, enabling the model to efficiently transmit and process information.

[0116] In a specific example, by using the set probability model, the connection between the visual area and the primary visual area in the heterogeneous brain model is analyzed. For example, the corresponding activation weight can be allocated between the output neural unit in the visual area and the input neural unit in the primary visual area: the connection weight between each visual neural unit and the primary visual neural unit may be set to 0.5. In this setting, the activation weight characterizes the connection strength and information transmission efficiency between the two data processing areas. After performing according to the above execution process, the obtained activation weight ensures that the model can efficiently transmit and process information during operation, thereby improving the overall performance of the heterogeneous brain model.

[0117] Step 404: Train the connection strength to obtain a trained heterogeneous brain model.

[0118] The method shown in this step has been described in step 302 and will not be elaborated here.

[0119] In summary, in the embodiment of the present application, by using the input feature data generated by acquiring driving sampling data and generating input feature data as the initial input of the heterogeneous brain model, the model can make full use of environmental perception data for efficient and intelligent obstacle avoidance; furthermore, by inputting the input feature data into each neuron unit in the heterogeneous brain model, the initialization of the model is completed, so that each neuron unit of the model is in a ready state, and the initialized neuron unit can accurately reflect the characteristics of the input data; then, by running the heterogeneous brain model, the running process of the data can be interpreted by using the corresponding relationship between the neuron units in the heterogeneous brain model. While realizing real-time obstacle avoidance on a low-power small embedded device, it also overcomes the high computational complexity and "black box" property of the deep convolutional neural network, realizes efficient computing under low-power conditions, and provides an intuitive explanation of the decision-making process; furthermore, by determining the output state data of the target neuron unit, the navigation state of the unmanned system is controlled, and an intelligent obstacle avoidance decision for the unmanned system is realized, ensuring efficient obstacle avoidance in a complex and changeable environment. Thus, based on the method of the embodiment of the present application, efficient and intelligent obstacle avoidance is realized, and the safety and reliability of the unmanned system are improved.

[0120] As Figure 9 shown, the embodiment of the present application also discloses a high-efficiency and intelligent obstacle avoidance device 50 for an unmanned system based on a heterogeneous brain model, which is applied to a controller of the unmanned system and includes:

[0121] An acquisition module 501, configured to acquire driving sampling data collected from sensors of the unmanned system, and generate input feature data for inputting into the heterogeneous brain model; the heterogeneous brain model includes a plurality of neuron units; each input neuron unit in the heterogeneous brain model has a corresponding output neuron unit; each input neuron unit is respectively used to acquire the state value of the output neuron unit corresponding to the input neuron unit, and solve the state value of the input neuron unit according to the acquired state value;

[0122] An input module 502, configured to input the input feature data into the heterogeneous brain model to initialize each neuron unit in the heterogeneous brain model;

[0123] An output module 503, configured to run the heterogeneous brain model, and determine the target output state data of the target neuron unit in the run heterogeneous brain model as the model output data of the heterogeneous brain model;

[0124] A control module 504, configured to control the navigation state of the unmanned system according to the model output data.

[0125] Optionally, the sensor is an image sensor, the driving sampling data is image driving sampling data, and the acquisition module 501 includes:

[0126] The downsampling sub-module is used to downsample the image driving sampling data to obtain multiple image block data of the image driving sampling data;

[0127] The feature sub-module is used to generate a data vector for each image block data, and determine the data matrix obtained by splicing the generated multiple data vectors as the input feature data.

[0128] Optionally, the control module 504 includes:

[0129] The component sub-module is used to obtain the value of the target data component in the model output data;

[0130] The control sub-module is used to control the navigation direction of the unmanned system according to the positive or negative situation of the value of the target data component, and control the navigation speed of the unmanned system according to the absolute value of the value of the target data component.

[0131] As Figure 10 shown, the embodiment of the present application also discloses a training device 60 for a heterogeneous brain model for efficient intelligent obstacle avoidance, which is used to train the heterogeneous brain model in the above embodiment of the method for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model, including:

[0132] The initialization module 601 is used to divide the heterogeneous brain model into multiple data processing areas, so that each neuron in the heterogeneous brain model has a corresponding data processing area, and establish a connection relationship between the neurons of every two data processing areas according to the initialized connection strength; the connection strength is used to characterize the connection quantity and connection weight between the multiple neurons included in the two data processing areas;

[0133] The training module 602 is used to train the connection strength to obtain a trained heterogeneous brain model.

[0134] Optionally, the initialization module 601 includes:

[0135] The division sub-module is used to divide the heterogeneous brain model into multiple data processing areas according to the number of neurons corresponding to each data processing area; there is a corresponding connection quantity and probability model between each data processing area and the target data processing area;

[0136] The connection sub-module is used to use each neuron in the data processing area as an output neuron, and determine multiple target neurons in the target data processing area as the input neurons of the output neuron according to the probability model and connection quantity between the data processing area and the target data processing area, and establish a data connection between the output neuron and the input neuron;

[0137] An activation sub-module is configured to establish an activation weight between the output data of a corresponding output neuron and the input data of an input neuron according to a probability model corresponding to the data connection between the corresponding output neuron and the input neuron.

[0138] In summary, in the embodiment of the present application, by using the input feature data generated by acquiring driving sampling data and generating input feature data as the initial input of the heterogeneous brain model, the model can make full use of environmental perception data for efficient and intelligent obstacle avoidance. Furthermore, by inputting the input feature data into each neuron in the heterogeneous brain model, the initialization of the model is completed, so that each neuron in the model is in a ready state, and the initialized neurons can accurately reflect the characteristics of the input data. Then, by running the heterogeneous brain model, the operation process of the data can be made interpretable by using the corresponding relationship between the neurons in the heterogeneous brain model. While achieving real-time obstacle avoidance on a low-power small embedded device, it also overcomes the high computational complexity and "black box" property of the deep convolutional neural network, realizes efficient computing under low-power conditions, and provides an intuitive explanation of the decision-making process. Furthermore, by determining the output state data of the target neuron, the navigation state of the unmanned system is controlled, and an intelligent obstacle avoidance decision for the unmanned system is realized, ensuring efficient obstacle avoidance in a complex and changeable environment. Thus, based on the method of the embodiment of the present application, efficient and intelligent obstacle avoidance is realized, and the safety and reliability of the unmanned system are improved.

[0139] The embodiment of the present application also provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, it implements each process of the above-mentioned method for efficient and intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model or the training method embodiment of a heterogeneous brain model for efficient and intelligent obstacle avoidance, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. Among them, the computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0140] Figure 11 It is a block diagram of an electronic device 700 provided by an embodiment of the present application. For example, the electronic device 700 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0141] Refer to Figure 11, the electronic device 700 may include one or more of the following components: a processing component 702, a memory 704, a power component 706, a multimedia component 708, an audio component 710, an input / output (I / O) interface 712, a sensor component 714, and a communication component 716.

[0142] The processing component 702 generally controls the overall operation of the electronic device 700, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to complete all or part of the steps of the above-mentioned efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model or the training method of a heterogeneous brain model for efficient intelligent obstacle avoidance. In addition, the processing component 702 may include one or more modules to facilitate the interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate the interaction between the multimedia component 708 and the processing component 702.

[0143] The memory 704 is used to store various types of data to support the operation of the electronic device 700. Examples of such data include instructions for any application or method operating on the electronic device 700, contact data, phone book data, messages, pictures, multimedia, etc. The memory 704 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0144] The power component 706 provides power to various components of the electronic device 700. The power component 706 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the electronic device 700.

[0145] The multimedia component 708 includes a screen that provides an output interface between the electronic device 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of touch or swipe actions but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 708 includes a front camera and / or a rear camera. When the electronic device 700 is in an operation mode, such as a shooting mode or a multimedia mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0146] The audio component 710 is used to output and / or input audio signals. For example, the audio component 710 includes a microphone (MIC) that is used to receive external audio signals when the electronic device 700 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 704 or transmitted via the communication component 716. In some embodiments, the audio component 710 further includes a speaker for outputting audio signals.

[0147] The I / O interface 712 provides an interface between the processing component 702 and a peripheral interface module, and the peripheral interface module can be a keyboard, a click wheel, buttons, etc. These buttons can include but are not limited to: a home button, a volume button, a power button, and a lock button.

[0148] The sensor component 714 includes one or more sensors for providing status assessments of various aspects of the electronic device 700. For example, the sensor component 714 can detect the on / off state of the electronic device 700, the relative positioning of components, such as the display and the keypad of the electronic device 700. The sensor component 714 can also detect a change in the position of the electronic device 700 or a component of the electronic device 700, the presence or absence of user contact with the electronic device 700, the orientation or acceleration / deceleration of the electronic device 700, and the temperature change of the electronic device 700. The sensor component 714 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor component 714 can also include a light sensor, such as a CMOS or a CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 714 can further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0149] The communication component 716 is used to facilitate communication between the electronic device 700 and other devices in a wired or wireless manner. The electronic device 700 can access a communication standard-based wireless network, such as WiFi, a carrier network (such as 2G, 3G, 4G, or 7G), or a combination thereof. In an exemplary embodiment, the communication component 716 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 716 further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0150] In an exemplary embodiment, the electronic device 700 can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components, for implementing the efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model provided in the embodiments of the present application or the training method for a heterogeneous brain model for efficient intelligent obstacle avoidance.

[0151] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions, and the above instructions can be executed by a processor 720 of the electronic device 700 to complete the above-mentioned efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model or the training method for a heterogeneous brain model for efficient intelligent obstacle avoidance. For example, the non-transitory storage medium can be a ROM, Random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0152] Figure 12 is a block diagram of an electronic device 800 shown according to an exemplary embodiment. For example, the electronic device 800 can be provided as a server. Referring to Figure 12 , the electronic device 800 includes a processing component 822, which further includes one or more processors, and memory resources represented by a memory 832 for storing instructions executable by the processing component 822, such as application programs. The application programs stored in the memory 832 can include one or more modules each corresponding to a set of instructions. In addition, the processing component 822 is configured to execute instructions to perform the efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model provided in the embodiments of the present application or the training method for a heterogeneous brain model for efficient intelligent obstacle avoidance.

[0153] The electronic device 800 may further include a power supply component 826 configured to perform power management of the electronic device 800, a wired or wireless network interface 850 configured to connect the electronic device 800 to a network, and an input / output (I / O) interface 858. The electronic device 800 may operate based on an operating system stored in the memory 832, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSD TM or the like.

[0154] The embodiments of the present application further provide a computer program product, including a computer program, which when executed by a processor, implements an efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model or a training method for a heterogeneous brain model for efficient intelligent obstacle avoidance.

[0155] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0156] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

[0157] Each embodiment in this specification is described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0158] It is easy for those skilled in the art to think that any combination application of the above various embodiments is feasible. Therefore, any combination of the above various embodiments is an embodiment of the present application. However, due to space limitations, this specification will not elaborate on each of them here.

[0159] The efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model or the training method for a heterogeneous brain model for efficient intelligent obstacle avoidance provided herein is not inherently related to any specific computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. It is obvious from the above description that the structures required to construct a system with the solution of the present application are obvious. In addition, the present application is not directed to any specific programming language. It should be understood that the content of the present application described herein can be implemented using various programming languages, and the description of the specific language above is to disclose the best implementation manner of the present application.

[0160] In the specification provided herein, a number of specific details are set forth. It will be appreciated, however, that embodiments of the present application may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail so as not to obscure an understanding of this description.

[0161] Similarly, it should be understood that in order to streamline the present application and assist in understanding one or more of the various aspects of the application, in the above description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected by the claims, the aspects of the application lie in less than all the features of the single embodiment disclosed above. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim stands on its own as a separate embodiment of the present application.

[0162] Those skilled in the art will appreciate that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all the features disclosed in this specification (including the accompanying claims, abstract and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract and drawings) can be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0163] In addition, those skilled in the art will be able to understand that although some of the embodiments described herein include certain features included in other embodiments rather than other features, the combination of the features of different embodiments means that it is within the scope of the present application and forms different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0164] Each component embodiment of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the method for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model or the training method of a heterogeneous brain model for efficient intelligent obstacle avoidance according to the embodiments of the present application. The present application can also be implemented as a device or apparatus program (for example, a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing the present application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0165] In another embodiment provided by the present invention, there is also provided a computer program product containing instructions, which when running on a computer, causes the computer to execute the method for efficient intelligent obstacle avoidance of an unmanned system based on a heterogeneous brain model or the training method of a heterogeneous brain model for efficient intelligent obstacle avoidance according to the embodiments of the present application.

[0166] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described according to the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a server, or a data center to another website, a computer, a server, or a data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

[0167] It should be noted that the above embodiments are illustrative of the present application rather than restrictive of the present application, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present application can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim listing several devices, several of these devices may be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words may be interpreted as names.

[0168] It should be noted that, for the method embodiments of the present application, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential for the embodiments of the present application.

[0169] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the system or device, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0170] The above is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are all included in the protection scope of the present invention.

Claims

1. An efficient intelligent obstacle avoidance method for unmanned systems based on a heterogeneous brain model, characterized in that, A controller applied to an unmanned system, comprising: Obtain driving sampling data collected from sensors of the unmanned system, and generate input feature data for inputting into a heterogeneous brain model according to the driving sampling data; the heterogeneous brain model includes a plurality of neural units; each input neural unit in the heterogeneous brain model has a corresponding output neural unit; each of the input neural units is used to obtain the state value of the output neural unit corresponding to the input neural unit, and solve the state value of the input neural unit according to the obtained state value; Input the input feature data into the heterogeneous brain model to initialize each neural unit in the heterogeneous brain model; Run the heterogeneous brain model, and determine the target output state data of the target neural unit in the run heterogeneous brain model as the model output data of the heterogeneous brain model; Control the navigation state of the unmanned system according to the model output data; A corresponding second-order non-linear differential calculation model is preset in each of the neural units, and the second-order non-linear differential calculation model is represented by the following formula group: ； ； Among them, represents the state value of the i-th neural unit; represents the state value of the j-th neural unit; represents the recovery variable of the i-th neural unit; represents the input feature data; represents the parameter of the linear growth rate of the neural unit; represents the parameter of the non-linear effect of the heterogeneous brain model; represents the set of neural units connected to the i-th neural unit; represents the transmission information weight between the j-th neural unit and the i-th neural unit; represents the perception weight of the input feature data to the i-th neural unit; is the coupling parameter, indicating the degree of influence; represents the attenuation parameter, controlling the recovery rate; represents the bias term of the heterogeneous brain model.

2. The efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model according to claim 1, characterized in that, The sensor is an image sensor, the driving sampling data is image driving sampling data, and the obtaining of the driving sampling data collected from the sensors of the unmanned system and generating the input feature data for inputting into the heterogeneous brain model according to the driving sampling data includes: Downsample the image driving sampling data to obtain a plurality of image block data of the image driving sampling data; Generate a data vector for each of the image block data, and determine the data matrix obtained by splicing the generated plurality of data vectors as the input feature data.

3. The efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model according to claim 1, wherein, The solving of the state value of the input neural unit according to the obtained state value includes: Execute iteration of the following formula group according to a preset number of iterations to solve the state value of the input neural unit: ； ； ； ； Among them, r and s represent intermediate variables of the algorithm, indicating the step size of iterative solution, indicating the moment corresponding to each state value; indicating at the moment, the state value of the input neuron, indicating at the moment, the recovery variable of the input neuron.

4. A training method for a heterogeneous brain model for efficient and intelligent obstacle avoidance, characterized in that, A method for training a heterogeneous brain model in the efficient intelligent obstacle avoidance method of an unmanned system based on a heterogeneous brain model according to any one of claims 1 to 3, comprising: Divide the heterogeneous brain model into a plurality of data processing areas, so that each neural unit in the heterogeneous brain model has a corresponding data processing area, and establish a connection relationship between the neural units of every two data processing areas according to the initialized connection strength; the connection strength is used to characterize the connection quantity and connection weight between the multiple neural units included in the two data processing areas; Train the connection strength to obtain a trained heterogeneous brain model.

5. The training method of the heterogeneous brain model for efficient intelligent obstacle avoidance according to claim 4, wherein The dividing of the heterogeneous brain model into a plurality of data processing areas, so that each neural unit in the heterogeneous brain model has a corresponding data processing area, and establishing a connection relationship between the neural units of every two data processing areas according to the initialized connection strength includes: Divide the heterogeneous brain model into a plurality of data processing areas according to the number of neural units corresponding to each data processing area; there is a corresponding connection quantity and probability model between each data processing area and the target data processing area; Take each neuron in the data processing area as an output neuron, and determine multiple target neurons in the target data processing area as the input neurons of the output neuron according to the probability model and the connection number between the target data processing areas, and establish a data connection between the output neuron and the input neuron; According to the probability model corresponding to the data connection between the corresponding output neuron and the input neuron, establish the activation weight between the output data of the corresponding output neuron and the input data of the input neuron.

6. An efficient intelligent obstacle avoidance device for an unmanned system based on a heterogeneous brain model, characterized in that, Applied to the controller of an unmanned system, including: An acquisition module, configured to acquire driving sampling data collected from sensors of the unmanned system, and generate input feature data for inputting into a heterogeneous brain model according to the driving sampling data; the heterogeneous brain model includes multiple neurons; each input neuron in the heterogeneous brain model has a corresponding output neuron; each input neuron is respectively used to obtain the state value of the output neuron corresponding to the input neuron, and solve the state value of the input neuron according to the obtained state value; An input module, configured to input the input feature data into the heterogeneous brain model to initialize each neuron in the heterogeneous brain model; An output module, configured to run the heterogeneous brain model, and determine the target output state data of the target neurons in the run heterogeneous brain model as the model output data of the heterogeneous brain model; A control module, configured to control the navigation state of the unmanned system according to the model output data; A corresponding second-order nonlinear differential calculation model is preset in each neuron, and the second-order nonlinear differential calculation model is characterized by the following formula group: ； ； Among them, represents the state value of the i-th neuron; represents the state value of the j-th neuron; represents the recovery variable of the i-th neuron; represents the input feature data; represents the parameter of the linear growth rate of the neuron; represents the parameter of the non-linear effect of the heterogeneous brain model; represents the set of neurons connected to the i-th neuron; represents the transmission information weight between the j-th neuron and the i-th neuron; represents the perception weight of the input feature data to the i-th neuron; is the coupling parameter, indicating the degree of influence; represents the decay parameter, controlling the recovery rate; represents the bias term of the heterogeneous brain model.

7. A training device for a heterogeneous brain model for efficient and intelligent obstacle avoidance, characterized in that, A method for training a heterogeneous brain model in the efficient intelligent obstacle avoidance method for an unmanned system based on a heterogeneous brain model according to any one of claims 1 to 3, including: An initialization module, configured to divide the heterogeneous brain model into multiple data processing areas, so that each neuron in the heterogeneous brain model has a corresponding data processing area, and establish a connection relationship between the neurons of every two data processing areas according to the initialized connection strength; the connection strength is used to characterize the connection number and connection weight between multiple neurons included in two data processing areas; A training module, configured to train the connection strength to obtain a trained heterogeneous brain model.

8. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

9. An electronic device, characterized in that, Including a processor, a memory, and a computer program stored on the memory and executable on the processor, and when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images