Neural network algorithm unmanned position intelligent sentry control system
Through the neural network algorithm control system, combined with the two-body sentinel guide rail and neural network unit, the problem of slow response speed and large error in the existing technology of single PID control is solved, and a high accuracy and fast response unmanned position smart sentinel is realized Control system improves safety and control accuracy.
Patent Information
- Application Number
- CN202410990281.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-08
AI Technical Summary
The existing unmanned sentinel control system at the position is mainly single-body PID control, with slow response speed and large errors, making it difficult to adapt to complex two-body control needs.
The neural network algorithm control system is adopted, and the first and second sentries are installed through the double-body sentry guide rail. Both are connected wirelessly with the neural network algorithm controller. The cross-connection between the neural network unit and the PID controller is used to realize the environmental data analysis of the target area. and safety level judgment, and perception and monitoring through infrared laser thermal imager and photoelectric tracking feedbacker, and use acoustic and optical weapon actuators to prevent and strike.
It improves the response speed and accuracy of unmanned sentry control, has the safety advantages of two-body back-to-back, can quickly identify and deal with abnormal situations in the target area, and provides high-precision prevention and strike capabilities.
Smart Images

Figure CN118915424A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent control technology, and in particular to a neural network algorithm unmanned position intelligent sentinel control system. Background Art
[0002] At present, most of the control systems for unmanned sentry positions at home and abroad are single-body, PID control. Although the control system is simple and easy to operate, it relies on precise control models, has slow control response speed, large errors, and is difficult to adapt to modular complex dual-body control. Summary of the invention
[0003] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a neural network algorithm unmanned position intelligent sentry control system, which solves the problems in the prior art that the patrol range of a single unmanned sentry in a traditional position is single (only inside or outside), the control system PID control needs to rely on an accurate control model, the control response speed is slow, the error is large, and it is difficult to adapt to modular complex dual-body control.
[0004] To achieve the above object, the present invention provides the following solutions:
[0005] A neural network algorithm unmanned position intelligent sentinel control system, comprising:
[0006] A double-body sentinel guide rail, a neural network algorithm controller, and a first sentinel and a second sentinel with identical structures;
[0007] The first sentinel and the second sentinel are respectively installed inside and outside the double-body sentinel guide rail in a back-to-back manner, and the first sentinel and the second sentinel are both wirelessly connected to the neural network algorithm controller;
[0008] The double-body sentinel guide rail is used to realize the defensive movement of the first sentinel and the second sentinel. The first sentinel and the second sentinel are respectively used to sense and monitor the inside and outside of the target area, obtain the environmental data of the target area, and alarm, prevent, drive away and attack abnormal conditions and suspicious objects in the target area. The neural network algorithm controller is used to analyze and judge suspicious objects in the target area according to the environmental data to obtain abnormal conditions and safety levels of the target area.
[0009] Preferably, the first sentinel comprises:
[0010] An environmental perception unit, an alarm, and a weapon unit, all connected to the neural network algorithm controller;
[0011] The environmental sensing unit is used to obtain environmental data of the target area, the alarm is used to issue an alarm, and the weapon unit is used to prevent, drive away and attack abnormal situations of suspicious objects in the target area.
[0012] Preferably, the environment sensing unit comprises:
[0013] An infrared laser thermal imager integrated feedback device and a photoelectric tracking feedback device are both connected to the neural network algorithm controller.
[0014] Preferably, the weapon unit comprises:
[0015] Acousto-optic weapon actuators, non-lethal weapon actuators, and height and direction servo actuators, all connected to a neural network algorithm controller.
[0016] Preferably, the neural network algorithm controller comprises:
[0017] Neural network unit and PID controller;
[0018] The PID controller is cross-connected with neurons in each layer of the neural network unit.
[0019] Preferably, the security level includes:
[0020] Safe, low risk, and high risk.
[0021] The present invention discloses the following technical effects:
[0022] The present invention provides a neural network algorithm unmanned position intelligent sentinel control system, comprising: a double-body sentinel guide rail, a neural network algorithm controller and a first sentinel and a second sentinel with the same structure; the first sentinel and the second sentinel are respectively installed inside and outside the double-body sentinel guide rail in a back-to-back manner, and the first sentinel and the second sentinel are both wirelessly connected to the neural network algorithm controller; the double-body sentinel guide rail is used to realize the defense movement of the first sentinel and the second sentinel, and the first sentinel and the second sentinel are respectively used to track, warn, prevent, drive away and strike suspicious objects inside and outside the target area, obtain environmental data of the target area, and track, warn, prevent, drive away and strike abnormal conditions in the target area, and the neural network algorithm controller is used to analyze and judge the target area according to the environmental data to obtain abnormal conditions and safety levels of the target area. The present invention has the advantages of double-body back-to-back safety superiority over the existing single body, and the neural network control of the double-body unmanned sentinel has fast response speed, small error and high control accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0024] Figure 1 A schematic diagram of the structure of an unmanned position intelligent sentinel control system using a neural network algorithm provided by an embodiment of the present invention;
[0025] Figure 2 A guide rail diagram and a moving schematic diagram of a double-body sentinel provided in an embodiment of the present invention;
[0026] Figure 3 A detailed schematic diagram of the unmanned intelligent sentinel control system for a neural network algorithm position provided by an embodiment of the present invention;
[0027] Figure 4 A structural diagram of a neural network controller provided by an embodiment of the present invention;
[0028] Figure 5 A schematic diagram of a control system provided by an embodiment of the present invention;
[0029] Figure 6 A schematic diagram of a neural network algorithm control operation process provided by an embodiment of the present invention;
[0030] Description of reference numerals:
[0031] 1-neural network algorithm controller, 2-first sentinel, 3-second sentinel, 4-double-body sentinel guide rail. DETAILED DESCRIPTION
[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0034] like Figure 1 As shown, the present invention provides a neural network algorithm unmanned position intelligent sentinel control system, comprising:
[0035] A double-body sentinel guide rail 4, a neural network algorithm controller 1, and a first sentinel 2 and a second sentinel 3 having the same structure;
[0036] The first sentinel 2 and the second sentinel 3 are installed inside and outside the double-body sentinel guide rail 4 in a back-to-back manner, and the first sentinel 2 and the second sentinel 3 are both wirelessly connected to the neural network algorithm controller 1;
[0037] The double-body sentinel guide rail 4 is used to realize the movement of the first sentinel 2 and the second sentinel 3. The first sentinel 2 and the second sentinel 3 are respectively used to sense, track and monitor the inside and outside of the target area, obtain the environmental data of the target area, and warn, prevent, drive away and strike against abnormal conditions in the target area. The neural network algorithm controller 1 is used to analyze and judge the target area according to the environmental data to obtain the abnormal conditions and safety level of the target area.
[0038] Specifically, suspicious objects include: people, vehicles, low-altitude aircraft, etc.
[0039] Specifically, the first sentinel (sentinel 1) 2 and the second sentinel (sentinel 2) 3 move synchronously inside and outside the guide rail in a back-to-back manner, as shown in the schematic diagram. Figure 2-3 As shown, the sentry is moved using a pulley system that can patrol a specific area inside the position through a preset path. The pulley system should have stability and precise control capabilities to ensure that the sentry can accurately reach the designated location. The range of movement of the sentry is carefully planned and considered for safety. The guide rail is a double-sided guide rail.
[0040] Furthermore, the first sentinel 2 includes:
[0041] An environmental perception unit, an alarm and a weapon unit, all connected to the neural network algorithm controller 1;
[0042] The environmental sensing unit is used to obtain environmental data of the target area, the alarm is used to issue an alarm, and the weapon unit is used to prevent abnormal situations in the target area.
[0043] Furthermore, the environment perception unit includes:
[0044] An infrared laser thermal imager integrated feedback device and a photoelectric tracking feedback device are both connected to the neural network algorithm controller 1.
[0045] Specifically, a 720° high-definition optoelectronic tracking camera is installed on the sentinel, and infrared, laser, and thermal imagers are integrated to sense and monitor the areas that need to be monitored, realizing multi-sensor fusion. Information collection is operated and calculated by the PID and neural network units of the neural network control system. Figure 4The network model regards each sensor as a neuron, and the connection between each neuron is a weight. All neurons are fused to form a threshold parameter, which is multiplied by the linear excitation function. The summation operation obtains the fused information processing to transmit the multi-sensor information of environmental perception in real time.
[0046] Furthermore, the weapon unit comprises:
[0047] The sound and light weapon actuator, the non-lethal weapon actuator and the height and direction servo actuator are all connected to the neural network algorithm controller 1.
[0048] Specifically, sentries carry sound and light alarms, sound and light repellents, and non-lethal weapons to respond to emergencies or potential threats approaching the position. The use and deployment of weapons should comply with relevant laws and regulations and be strictly managed and controlled.
[0049] Furthermore, the neural network algorithm controller 1 comprises:
[0050] Neural network unit and PID controller;
[0051] The PID controller is cross-connected with neurons in each layer of the neural network unit.
[0052] Specifically, the neural network algorithm controller 1 is composed of a neural network algorithm and a PID controller. Its structural design is based on the neural network structure, and is composed of an input layer, a hidden layer, and an output layer. Its structure is composed of multiple single neurons and PID controllers that are cross-composed. The specific structure diagram of the controller is shown in FIG. Figure 4 The neural network algorithm controller 1 can realize accurate real-time control of the sound and light weapon actuator, non-lethal weapon actuator, height and direction servo actuator, infrared laser thermal imager integrated feedback device, and optoelectronic tracking feedback device. Its schematic diagram is shown in Figure 5 shown.
[0053] The specific control principle of the neural network algorithm controller 1 is as follows:
[0054] Depend on Figure 4 In the neural network algorithm structure, the acoustic and optical weapon actuator, non-lethal weapon actuator, high and low and direction servo actuator, infrared laser thermal imager integrated feedback device, and optoelectronic tracking feedback device can be equivalent to the neurons in the hidden layer of the neural network algorithm, and the neural network algorithm receives the neuron setting value of the input signal from the input layer neuron, passes through the amplifier, and passes through the connection weight parameters between neurons, and is combined with the unit and the excitation function operation, such as Figure 6 As shown, the control of the neural network algorithm is realized.
[0055] Control steps:
[0056] (1) Construct a suitable neural network model structure, determine the network input layer nodes, output layer nodes and hidden layer nodes, and set the weight coefficients as the initial values of each layer of the network to determine the appropriate values of the network learning rate and inertia coefficient according to the actual situation.
[0057] (2) The network can adopt y in (k), y out (k), calculate e(k)=y in (k)-y out (k);
[0058] y in (k) represents the neural network input value; y out (k) represents the output value of the neural network; e(k) represents the error of the neural network;
[0059] (3) Determine the input O1 of the neural network (1) =e(k),O2 (1) =e(k-1),O3 (1) =e(k-2);
[0060] O1 (1) Represents the first layer error; O2 (1) Represents the second layer error; O3 (1) Represents the layer 3 error.
[0061] (4) Determine u(k) of the PID controller network;
[0062] u(k) represents the input of the neural network controlled object;
[0063] (5) Determine the input and output of each layer of the neural network algorithm PID network;
[0064] (6) The output layer of the neural network algorithm PID network corrects the weight coefficient value;
[0065] (7) Determine the hidden layer correction weight coefficient value of the neural network algorithm PID network;
[0066] (8) Next, take k = k + 1, return to the first step, let the network sample again, and repeat the process again; k represents the calculation to k steps until the network weights converge, and then return;
[0067] The neural network algorithm uses its own learning and optimization characteristics to achieve real-time self-adjustment of PID controller parameters and improve system control accuracy.
[0068] Furthermore, the security level includes:
[0069] Safe, low risk, and high risk.
[0070] Specifically, the data collected by the sentinel system and the multi-sensor information are processed. The received data is used as the data source, and the gradient warning method is used for intelligent self-analysis. The response is made in the form of encrypted response, and the security threshold is set to analyze the risk level and degree of the information. Monitor whether the data change range deviates from the overall statistical characteristics of the security threshold time series. If it exceeds a certain range value, it can be considered that there is an abnormal situation. According to the deviation, different levels of warning can be issued, which are divided into three levels: safe, low risk, and high risk. The system should be able to analyze and identify abnormal behavior in real time, and make autonomous decisions according to the three levels of danger.
[0071] Specifically, the Sentinel system has protective measures to prevent local sabotage or manipulation. Key components such as control systems and weaponry should be designed to be waterproof, shockproof and vandal-proof. The system has strict identity authentication and access control mechanisms to ensure that only authorized personnel can access and manipulate the system.
[0072] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0073] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of the present invention. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A neural network algorithm unmanned position intelligent sentinel control system, characterized in that: include: A double-body sentinel guide rail, a neural network algorithm controller, and a first sentinel and a second sentinel with identical structures; The first sentinel and the second sentinel are respectively installed inside and outside the double-body sentinel guide rail in a back-to-back manner, and the first sentinel and the second sentinel are both wirelessly connected to the neural network algorithm controller; The double-body sentinel guide rail is used to realize the defensive movement of the first sentinel and the second sentinel. The first sentinel and the second sentinel are respectively used to sense and monitor the inside and outside of the target area, obtain the environmental data of the target area, and warn, drive away and strike against abnormal conditions in the target area. The neural network algorithm controller is used to analyze and judge the target area according to the environmental data to obtain the abnormal conditions and safety level of the target area.
2. According to the neural network algorithm unmanned position intelligent sentinel control system of claim 1, it is characterized in that: The first sentinel includes: An environmental perception unit, an alarm, and a weapon unit, all connected to the neural network algorithm controller; The environmental perception unit is used to obtain environmental data of suspicious objects in the target area, the alarm is used to issue a warning alarm, and the weapon unit is used to prevent, drive away and attack abnormal situations in the target area.
3. According to claim 2, a neural network algorithm unmanned position intelligent sentinel control system is characterized in that: The environment sensing unit comprises: An infrared laser thermal imager integrated feedback device and a photoelectric tracking feedback device are both connected to the neural network algorithm controller.
4. According to claim 2, a neural network algorithm unmanned position intelligent sentinel control system is characterized in that: The weapon unit comprises: Acousto-optic weapon actuators, non-lethal weapon actuators, and height and direction servo actuators, all connected to a neural network algorithm controller.
5. According to the neural network algorithm unmanned position intelligent sentinel control system of claim 1, it is characterized in that: The neural network algorithm controller comprises: Neural network unit and PID controller; The PID controller is cross-connected with neurons in each layer of the neural network unit.
6. According to the neural network algorithm unmanned position intelligent sentinel control system of claim 1, it is characterized in that: The security levels include: Safe, low risk, and high risk.
