Mouse autonomous navigation behavior closed-loop regulation and control device and method based on nerve regulation and control

Abstract

The invention discloses a mouse autonomous navigation behavior closed-loop regulation and control device and method based on nerve regulation and control. Based on the combination of an Arduino IDE development platform and a multi-channel biological signal conditioning module, an intelligent labyrinth experimental platform with environmental adaptability is constructed, and real-time monitoring and closed-loop regulation and control of autonomous navigation behaviors of experimental animals in a complex space environment are realized. The device effectively overcomes the technical defects of low time sequence control precision, limited system expansibility, too high experiment cost and the like of traditional labyrinth equipment, and the operation convenience, economical efficiency and environment adaptability of the system are remarkably improved. The device can synchronously collect multi-modal behavioral parameters such as a space trajectory map, decision incubation period data and a path optimization strategy, and provides a programmable closed-loop regulation and control innovative method system for neural circuit function analysis by establishing a dynamic correlation model of behavioral characteristics and neural regulation and control parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of neurobehavioral regulation, and in particular to a device and method for closed-loop regulation of autonomous navigation behavior of mice based on neuroregulation. Background Art

[0002] With the rapid development of neuroethological research, the demand for real-time collection and precise control of behavioral data continues to increase. Because traditional mazes often utilize fixed physical structures and cannot dynamically reconfigure environmental topology, this severely restricts the research dimension of spatial cognitive plasticity. Providing higher-performance solutions has become a necessity. Existing behavioral feedback devices generally suffer from excessive stimulation latency, and the insufficient sensitivity of mechanical touch detection modules leads to the loss of subtle behavioral characteristics of animals, making it impossible to support dynamic coding analysis of decision-making neural circuits. Furthermore, most systems utilize open control architectures and lack closed-loop synchronization mechanisms for behavior-environment-reward. This makes it impossible to effectively simulate real-time control processes and difficult to support dynamic coding analysis of decision-making neural circuits. Summary of the Invention

[0003] The purpose of this invention is to solve the problems existing in the prior art and provide a free maze exploration behavior decision-making device based on Arduino. The specific technical solution is as follows:

[0004] 1. A closed-loop control device for autonomous navigation behavior of mice based on neural regulation; comprising:

[0005] The core control unit is composed of an Arduino UNO R3 development board, which controls the reinforcement learning control module, which is composed of a pneumatic displacement platform, a water pump, and a buzzer.

[0006] The behavioral feedback execution module consists of multiple touch buttons, each of which is equipped with a tactile sensor, which captures the navigation selection signals of the experimental subject in real time through the tactile sensor;

[0007] A multimodal interaction module, including an optical projection device, generates a three-dimensional interactive maze topology in real time through a dynamic optical projection device (the three-dimensional interactive maze topology is realized based on a real-time two-dimensional path coordinate transformation algorithm);

[0008] Arduino IDE programmable control unit is connected to the core control unit through the signal input port.

[0009] The multiple touch buttons include left, right, and forward touch buttons, which drive the pneumatic displacement platform to move left, right, and forward, thereby driving the experimental object located on the pneumatic displacement platform to move left, right, and forward in the maze space.

[0010] The Arduino uno r3 development board is loaded with a program in the Arduino IDE programmable control unit, and a physical connection mapping of the pneumatic displacement platform, the water pump, and the buzzer is established based on the Arduino uno r3 development board.

[0011] The program of the Arduino IDE programmable control unit includes a pin allocation protocol, a timing control function, and a serial port data acquisition module; the pin allocation protocol is used to initialize the reinforcement learning control module; the timing control function is used to build a high-precision sequential logic controller to define the single trigger threshold and periodic training timing parameters of each actuator in the reinforcement learning control module; the serial port data acquisition module is used to integrate a serial communication monitoring interface to capture and store the state switching timing data of each actuator in real time.

[0012] 2. A closed-loop control method for autonomous navigation behavior in mice based on neural regulation

[0013] Step 1) Write the program of the Arduino IDE programmable control unit and enter the program into the Arduino Uno R3 development board;

[0014] Step 2) Building a closed-loop control experimental platform for autonomous navigation behavior of mice;

[0015] The step 2) is specifically as follows:

[0016] The Arduino IDE programmable control unit is connected to the core control unit via the signal input port. The optical projection system generates a three-dimensional spatial configuration in real time based on the two-dimensional maze topology data. The behavioral feedback execution module captures the experimental subject's selection signal in real time through a tactile sensor and inputs it into the core control unit through a digital signal isolation circuit. The core control unit activates the reinforcement learning mechanism and synchronously drives the pneumatic displacement platform to achieve dynamic adaptation of the experimental environment parameters. The reinforcement learning mechanism includes broadband white noise stimulation (duration 2 seconds) or a quantitative liquid delivery device.

[0017] A three-dimensional interactive maze topology is realized based on a real-time two-dimensional path coordinate transformation algorithm. The topology generates an interactive maze holographic environment in the maze space through an optical projection device.

[0018] Step 3) Based on the experimental platform constructed in step 2), a closed-loop control method for autonomous navigation behavior of mice based on neural regulation is constructed, including a training phase and a testing phase.

[0019] The training phase is specifically as follows:

[0020] A dynamic optical projection system generates a three-dimensional interactive maze topology in real time, providing sub-millimeter spatial visual stimulation. When the subject enters the preset maze space, the subject touches the touch button to drive the pneumatic displacement platform to move, thereby driving the subject to move within the maze space.

[0021] If the subject chooses the correct movement route, the core control unit activates the water pump to give a certain amount of liquid reward.

[0022] On the contrary, if the decision is wrong or the delay exceeds the 500ms threshold, the core control unit activates the buzzer and the pneumatic displacement platform returns to the position before the decision.

[0023] The testing phase is specifically as follows:

[0024] During the testing phase, a spatial mapping inversion mechanism was implemented, mirroring the coordinates of the starting and ending points during the training phase. A countdown trigger mechanism was used to set a 60-second time threshold. If the subject did not reach the target area (endpoint) within the window (60-second time threshold), a white noise stimulus was activated. Successful navigation triggered a quantitative liquid reward module. The dynamic trajectory data of the pneumatic platform was optimized for environmental adaptation accuracy in real time using a Kalman filter algorithm (i.e., ensuring consistent single-move distances).

[0025] During the testing phase, the activity signals of the motor cortex neurons of the experimental subjects were extracted simultaneously to provide data for subsequent research on the visual stimulation decision-making behavior of the experimental subjects.

[0026] Beneficial effects of the present invention:

[0027] The present invention constructs a closed-loop control system and method for autonomous navigation behavior in mice based on neural regulation. The core control unit is constructed using the Arduino microcontroller platform to achieve precise timing control of multimodal neural regulation strategies. Compared with traditional free behavior observation devices, this system achieves an order of magnitude improvement in the sensitivity of decision-making behavior detection by establishing a closed-loop feedback mechanism between behavioral response and neural stimulation. The system innovatively constructs a modular multimodal neural regulation system, which is scalable and compatible with optogenetic precision intervention and neural electrical signal synchronization acquisition modules. It establishes a dynamic correlation model of "environmental variables-behavioral output-signal analysis" at the neural circuit level, establishing an innovative research paradigm and technical platform for analyzing the collaborative coding mechanism of the prefrontal cortex and hippocampus in complex spatial cognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a structural schematic diagram of the closed-loop control device for autonomous navigation behavior of mice based on neural regulation provided by the present invention.

[0029] Figure 2 It is a flow chart of the closed-loop control method of autonomous navigation behavior of mice based on neural regulation provided by the present invention.

[0030] In the figure: Arduino IDE programmable control unit (1), signal input port (2), core control unit (3), optical projection device (4), maze space (5), pneumatic displacement platform (6), water pump (7), buzzer (8), left touch button (9), right touch button (10), forward touch button (11). DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described below with reference to the accompanying drawings and embodiments.

[0032] like Figure 1 Figure 1 shows the schematic structure of the neural-controlled closed-loop control device for autonomous navigation behavior in mice. Compared to traditional auditory stimulation devices, this device is not only simple to operate, low-cost, and highly flexible, but also provides a powerful tool for studying the neural mechanisms of free exploratory behavioral decision-making. It contributes to a deeper understanding of the brain's information processing and decision-making processes in complex environments. Furthermore, based on the openness and scalability of Arduino, this device can be combined with other technologies, such as brain-computer interfaces and neuroimaging, providing new ideas and methods for multimodal neuroscience research.

[0033] The present invention proposes a closed-loop control device for autonomous navigation behavior of mice based on neural regulation, which is composed of a core control unit (Arduino uno r3 development board), a multimodal interaction module and a behavior feedback execution module. The device comprises an Arduino IDE programmable control unit, a signal input port, a core control unit, an optical projection system, a pneumatic positioning platform, a micro water pump, a broadband acoustic stimulator, and a micro force sensing detection unit. The Arduino IDE programmable control unit is connected to the core control unit through a circuit, and the optical projection system generates a three-dimensional spatial configuration in real time based on two-dimensional maze topology data; the behavior feedback execution module captures navigation selection signals in real time through bilateral tactile sensors and inputs them into the core control unit through a digital signal isolation circuit. The reinforcement learning control module comprises a pneumatic displacement platform (6), a micro water pump (7) and a buzzer (8), and the reinforcement learning control module realizes multi-subsystem timing synchronization through a temperature-compensated crystal oscillator.

[0034] like Figure 2 As shown in FIG, a closed-loop control method for autonomous navigation behavior of mice based on neural regulation, the specific steps are as follows:

[0035] (a) Code development for a closed-loop neural control embedded synchronous triggering architecture: Configure a multi-channel hardware interface for a micro-force sensing unit, a microfluidic control unit, a broadband acoustic stimulator, an optical projection system, and a pneumatic positioning device; execute a modular initialization protocol to put each actuator into standby mode; construct a high-precision sequential logic controller to define the single trigger threshold and periodic training timing parameters; and integrate a serial communication monitoring interface to capture and store the state switching timing data of each actuator in real time.

[0036] (b) Construction of a closed-loop control experimental platform for the free behavior paradigm: Using an embedded programmable logic control system, the core computing unit is constructed based on the Arduino microcontroller platform, and a multimodal behavior control component cluster (including a micro-pump drive module, a broadband acoustic stimulation array, and a tactile mechanical sensing matrix) is integrated to establish a closed-loop feedback system for biological behavior-environment interaction. A dynamic optical projection system is used to generate a three-dimensional interactive maze topology in real time (the three-dimensional interactive maze topology is realized based on a real-time two-dimensional path coordinate transformation algorithm, and the topology is used to generate an interactive maze holographic environment in the maze space through an optical projection device (4)); a micro-force sensing detection unit is used to continuously capture the behavioral mechanical characteristics of the experimental subject. Based on the captured behavioral mechanical characteristics of the experimental subject, an associated reinforcement learning mechanism is activated. The reinforcement learning mechanism includes broadband white noise stimulation (duration 2s) or a quantitative liquid delivery device, and the pneumatic suspension platform is synchronously driven to achieve dynamic adaptation of the experimental environment parameters.

[0037] (c) A closed-loop control method for autonomous navigation behavior of mice based on neural regulation: It includes a dual-modal control module of training phase and testing phase: in the training phase, a dynamic optical projection system is used to generate a three-dimensional interactive maze topology structure in real time to provide submillimeter spatial visual stimulation. When the experimental subject enters the preset navigation space, a tactile sensor array is used to capture its path selection behavior. The core control unit (3) analyzes the decision signal through the digital signal processing interface and executes a triple reinforcement logic: when a correct spatial decision is detected, the volumetric micro-pump device, the pneumatic positioning platform forward vector excitation (displacement speed 5cm / s±5%) and the maze topology gradient reconstruction algorithm (based on the path cost function to update the weight matrix) are simultaneously activated; if the decision is wrong or the delay exceeds the 500ms threshold, the broadband noise excitation source and the pneumatic platform reverse displacement control (speed 3cm / s±8%) are triggered, and the maze parameters are locked (i.e., the maze environment is kept unchanged) until the experimental subject completes the path exploration. During the testing phase, a spatial mapping inversion mechanism is implemented, mirroring the coordinates of the starting and ending points of the training phase. A countdown trigger mechanism is used to set a 60-second time threshold. If the subject fails to reach the target area within the window period, white noise stimulation is activated. Successful navigation triggers a quantitative liquid reward module. The dynamic trajectory data of the pneumatic platform is optimized for environmental adaptation accuracy in real time using a Kalman filter algorithm.

Claims

1. A closed-loop control device for autonomous navigation behavior of mice based on neural regulation, characterized in that: include: A core control unit (3) is composed of an Arduino uno r3 development board, and controls a reinforcement learning control module through the Arduino uno r3 development board, wherein the reinforcement learning control module is composed of a pneumatic displacement platform (6), a water pump (7), and a buzzer (8); The behavioral feedback execution module consists of multiple touch buttons, each of which is equipped with a tactile sensor, which captures the navigation selection signals of the experimental subject in real time through the tactile sensor; A multimodal interaction module includes an optical projection device (4) for generating a three-dimensional interactive maze topology in real time through the dynamic optical projection device (4); The Arduino IDE programmable control unit (1) is connected to the core control unit (3) via a signal input port (2).

2. The closed-loop control device for autonomous navigation behavior of mice based on neural regulation according to claim 1, characterized in that: The plurality of touch buttons include left, right, and forward touch buttons, which drive the pneumatic displacement platform (6) to move left, right, and forward, thereby driving the experimental object located on the pneumatic displacement platform (6) to move left, right, and forward in the maze space.

3. The closed-loop control device for autonomous navigation behavior of mice based on neural regulation according to claim 1, characterized in that: The Arduino uno r3 development board is loaded with a program in the Arduino IDE programmable control unit (1), and a physical connection mapping of the pneumatic displacement platform (6), the water pump (7), and the buzzer (8) is established based on the Arduino uno r3 development board.

4. The closed-loop control device for autonomous navigation behavior of mice based on neural regulation according to claim 3, characterized in that: The program of the ArduinoIDE programmable control unit (1) includes a pin allocation protocol, a timing control function, and a serial port data acquisition module; The pin allocation protocol is used to initialize the reinforcement learning control module; The timing control function is used to construct a high-precision timing logic controller and define the single trigger threshold and periodic training timing parameters of each actuator in the reinforcement learning control module; The serial port data acquisition module is used to integrate the serial communication monitoring interface to capture and store the state switching timing data of each actuator in real time.

5. A method for constructing a closed-loop control method for autonomous navigation behavior of mice based on neural regulation using the device according to any one of claims 1 to 4, characterized in that: 1) Write the program of the Arduino IDE programmable control unit (1) and input the program into the Arduino uno r3 development board; 2) Build an experimental platform for closed-loop control of autonomous navigation behavior in mice; 3) Based on the experimental platform constructed in step 2), a closed-loop control method for autonomous navigation behavior of mice based on neural regulation is constructed, including a training phase and a testing phase.

6. The closed-loop control method for autonomous navigation behavior of mice based on neural regulation according to claim 5, characterized in that: The step 2) is specifically as follows: The Arduino IDE programmable control unit (1) is connected to the core control unit (3) via the signal input port (2), and the optical projection system (4) generates a three-dimensional spatial configuration in real time based on the two-dimensional maze topology data; The behavior feedback execution module captures the selection signal of the experimental subject in real time through the tactile sensor and inputs it into the core control unit through the digital signal isolation circuit; the core control unit (3) activates the reinforcement learning mechanism and synchronously drives the pneumatic displacement platform (6) to achieve dynamic adaptation of the experimental environment parameters.

7. The closed-loop control method for autonomous navigation behavior of mice based on neural regulation according to claim 6, characterized in that: In step 3), the training phase is specifically as follows: A three-dimensional interactive maze topology is generated in real time through a dynamic optical projection system to provide sub-millimeter spatial visual stimulation. When the subject enters the preset maze space, the subject drives the pneumatic displacement platform (6) to move by touching the touch button, thereby driving the subject to move in the maze space. If the subject chooses the correct movement route, the core control unit (3) activates the water pump to give a fixed amount of liquid reward; On the contrary, if the decision is wrong or the delay exceeds the 500ms threshold, the core control unit (3) activates the buzzer and the pneumatic displacement platform (6) returns to the position before the decision.

8. The closed-loop control method for autonomous navigation behavior of mice based on neural regulation according to claim 6, characterized in that: In step 3), the testing phase is specifically as follows: During the testing phase, a spatial mapping inversion mechanism was implemented, mirroring the starting and ending coordinates of the training phase. A countdown trigger mechanism was used to set a 60-second time threshold. If the subject did not reach the target area within the window period, white noise stimulation was activated. Successful navigation triggered a quantitative liquid reward module. The dynamic trajectory data of the pneumatic platform was used to optimize the environmental adaptation accuracy in real time using a Kalman filter algorithm. During the testing phase, the activity signals of the motor cortex neurons of the experimental subjects were extracted simultaneously to provide data for subsequent research on the visual stimulation decision-making behavior of the experimental subjects.

Images