Mobile obstacle device capable of being remotely controlled by coach

Abstract

The invention belongs to the technical field of sports equipment, and discloses a coach remote control moving obstacle device which comprises a machine shell, universal wheels, a laser camera, a charging interface, a USB interface, a displayer, a controller, an AI voice device, a speed sensor, a wireless transceiver and a motor. Universal wheels are fixed to the two sides of the bottom of the machine shell through screws. A laser camera is embedded in the center of the front face of the shell. A coach can remotely control the moving obstacle device to establish contact with an athlete to move along with the athlete, and the athlete can get rid of defense only when the speed of the athlete exceeds the moving obstacle speed. And the running speed and distance of the athlete can be accurately measured through the laser camera. Meanwhile, an AI large model is integrated through the AI voice device, data analysis can be more intelligently carried out, data can be accurately processed, and a more professional exercise scheme is provided for athletes.

Description

Technical Field

[0001] The invention belongs to the technical field of sports equipment, and in particular relates to a trainer's remote control mobile obstacle device. Background Art

[0002] Obstacle courses typically consist of a series of fixed or movable obstacles designed to help users improve their flexibility, coordination, endurance, and ability to navigate complex environments. These courses are widely used in school physical education programs, corporate team building, and personal fitness. However, existing obstacle courses lack intelligence and the data they capture is inaccurate.

[0003] The closest existing technology is an intelligent training obstacle course device, consisting of a control module, movable wheels, basic sensors, and a remote control module. It is suitable for physical and reaction training. While this device can achieve a certain degree of automated movement and path adjustment, it relies primarily on simple remote control and sensor feedback, lacking more advanced human-computer interaction and environmental perception capabilities.

[0004] The problems with this technology are: first, the control method is single, and it only supports traditional remote control operations, and lacks more natural human-computer interaction functions such as voice control; second, the obstacle avoidance and recognition capabilities are limited, and high-precision visual modules such as laser cameras are not used, making it difficult to adapt to complex training scenarios; third, the lack of expansion interfaces and real-time display functions limits the system's flexibility and the coach's intuitive grasp of the equipment status, and the level of intelligence is still insufficient. Summary of the Invention

[0005] In view of the problems existing in the prior art, the present invention provides a mobile obstacle device that can be remotely controlled by a coach.

[0006] The present invention is achieved in that a coach can remotely control a mobile obstacle device comprising: Casing, universal wheel, laser camera, charging port, USB port, display, controller, AI voice device, speed sensor, wireless transmitter and receiver, motor; Universal wheels are fixed on both sides of the bottom of the casing by screws; a laser camera is embedded in the center of the front of the casing; a charging port is opened on the left right side of the casing; a USB port is opened on the right right side of the casing; a display is fixed in the center of the top of the casing by screws; and a controller is fixed in the center of the casing by screws.

[0007] Furthermore, an AI voice device is fixed to the lower left side of the casing by screws.

[0008] Furthermore, a speed sensor is fixed on the upper left side of the housing by screws.

[0009] Furthermore, a wireless transmitter and receiver is fixed on the upper right side of the housing by screws.

[0010] Furthermore, an electric motor is fixed at the lower right side of the casing by screws.

[0011] Furthermore, the controller is connected to the AI voice device, the speed sensor, the wireless transmitter and receiver, and the motor through circuit lines; Furthermore, the coach can remotely control the mobile obstacle using a smart belt that is connected to a wireless transmitter and receiver to establish contact with the athlete so that the mobile obstacle can follow.

[0012] Furthermore, the smart belt can simultaneously control three or more coaches to remotely move the obstacle course.

[0013] Another object of the present invention is to provide a trainer-controlled remote-controlled mobile obstacle control system based on the trainer-controlled remote-controlled mobile obstacle control system, the system comprising a central control unit, a wireless communication module, a voice recognition processor, a motor drive unit, and a multi-source sensor interface module, wherein: The central control unit is located in the middle area of the housing and is used to run the embedded control program to implement instruction parsing, state scheduling and path updating; The wireless communication module includes a transmitter and a receiver, which establishes a point-to-point connection with an external remote control terminal based on WiFi or Bluetooth communication protocol, and is used to receive control instructions issued by the coach and transmit them to the central control unit; The speech recognition processor is connected to the AI speech device, supports the local semantic parsing algorithm, and after recognizing the keywords, forwards the control intention to the central control unit in the intermediate instruction format; The motor drive unit is linked to the central control unit to adjust the motor speed through the PWM signal to achieve closed-loop control of the obstacle movement state; The multi-source sensor interface module is connected to the speed sensor, position encoder and attitude sensor to build a real-time state feedback path for the device and support adaptive correction of motion parameters based on the PID algorithm.

[0014] Furthermore, the central control unit adopts a microcontroller architecture based on the STM32 or ARM CortexM series, is embedded with a real-time operating system, and concurrently manages voice control, remote control response and speed closed-loop adjustment tasks using a task priority queue scheduling strategy.

[0015] Furthermore, the speech recognition processor integrates an end-side inference engine, uses a lightweight neural network model for keyword recognition, and performs asynchronous data interaction with the central control unit via serial communication or SPI bus; 1) The voice signal data processing process begins with the AI voice device collecting sound waves from the coach's commands. A MEMS microphone array is used for beamforming to improve the signal-to-noise ratio of the commands. The voice signal then enters the voice recognition processor for on-device inference. This processor, with a built-in lightweight neural network model, performs keyword classification and recognition through Mel-spectrogram extraction and feature vector quantization. The parsed control intent is encapsulated into a structured intermediate control frame and sent to the central control unit via the SPI or UART interface. The task scheduler then loads it into the command buffer queue for subsequent parsing and execution. 2) The remote control wireless command link establishes a low-latency point-to-point communication channel based on the WiFi Direct or BLE protocol stack. Control commands sent by the instructor's remote control are encapsulated in simplified control frames at the data link layer and sent to the system's wireless communication module via the GATT or UDP protocol. Upon receiving the frame data, the receiver performs header check, CRC check, and channel error correction, and then passes the payload to the central control unit for unpacking and parsing. Command content, including target speed, direction angle, or stop flag, is encoded in a standard fixed-length byte format to ensure real-time and stability. 3) Control command parsing and motor control signal generation are triggered by the central control unit's RTOS-based event-driven mechanism. The parsed control commands are sent via an internal bus to the motor drive unit, which converts the target speed into a duty cycle adjustment value and outputs a PWM signal to drive a brushed or brushless DC motor. The drive unit has a built-in H-bridge control circuit and current protection module to ensure output response time within milliseconds, improving overall machine response sensitivity. In addition, the system supports dynamic speed limiting and steering adjustment logic, dynamically setting the motor's target operating range through limit registers, enhancing control stability and safety margin control capabilities. 4) The multi-source state feedback processing mechanism is completed by the sensor fusion subsystem formed by the speed sensor, encoder, and attitude sensor. The speed sensor provides linear velocity estimation, the rotary encoder outputs pulse counts to convert angular displacement, and the attitude sensor outputs three-dimensional attitude angles through a combination of six-axis or nine-axis gyroscopes and accelerometers. All sensor data is uniformly sampled by the ADC or I2C / SPI transmission channel, and the state fusion algorithm built into the central control unit performs data integration. The final system uses an incremental PID regulator to compensate for the current error and dynamically adjust the PWM signal to achieve real-time closed-loop correction of the motion state and adaptive adjustment of the path deviation.

[0016] Another object of the present invention is to provide a control method based on the above-mentioned trainer-controlled remote control system for a mobile obstacle device, the method comprising: Step 1: Voice command collection and edge semantic processing; 1.1 The coach's voice commands are first collected by an AI voice device. The device's front-end uses a pickup system consisting of a micro-microphone array. This system uses a spatial filtering algorithm to enhance the forward voice signal while reducing ambient noise interference. 1.2 The pre-processed speech signal is fed into the speech recognition processor, where edge inference is performed using a locally deployed lightweight neural network model. The model first converts the raw audio signal into a Mel-spectrogram, extracts vectors representing semantic features, and finally inputs these vectors into the neural network for keyword classification and recognition. 1.3 Once a target keyword is identified with high confidence, the speech recognition processor converts the recognition result into an intermediate control frame and transmits it asynchronously to the central control unit via serial communication or SPI bus; Step 2: Remote control command link establishment and data analysis; 2.1 The instructor's handheld remote control terminal establishes a low-latency, point-to-point data channel with the device's wireless communication module via WiFi Direct or BLE low-power protocols. This channel features broadcast discovery, fast connection, and handshake confirmation mechanisms. 2.2 The control command is encapsulated into a concise data frame through the protocol, which includes the command type, target value, command timestamp and redundant bits for verification; 2.3 After the receiving wireless module completes the structural analysis and verification of the data frame, it unpacks the command content and forwards it to the command processing module of the central control unit, which adds it to the control command queue for execution; Step 3: Central control logic analysis and motor drive signal generation; 3.1 The central control unit is based on an embedded real-time operating system and uses a task scheduler to manage all input events, including voice commands, remote control input, and sensor feedback; 3.2 Whenever a new control command is added to the buffer queue, the system initiates the motor control task based on the command type, converting the received target speed and direction into a motor speed control signal. During this process, the central control unit calculates the PWM signal output parameters corresponding to the target motion state and sends them to the motor drive unit. 3.3 The motor drive unit includes a complete power regulation circuit and current protection mechanism, which can effectively prevent system instability caused by sudden load changes while ensuring responsiveness. In addition, dynamic speed and steering limit logic is provided to dynamically set the operating range based on the training environment or mission requirements, improving the control safety of the equipment. Step 4: Multi-source sensor feedback and closed-loop motion control; 4.1 The control system uses a sensor fusion path formed by speed sensors, encoders, and attitude sensors to perform high-frequency sampling and feedback on the real-time status of the equipment; The speed sensor is used to measure the actual operating speed of the equipment; Encoders detect changes in displacement or angle; The attitude sensor provides three-dimensional attitude information of the device, including parameters such as roll, pitch, and yaw; 4.2 All sensor data is uniformly transmitted to the control unit via the system's internal I²C or SPI interface, and is integrated using a fusion algorithm to filter out noise and fill in missing information; 4.3 The control unit determines the current motion state based on the fused data and compares it with the desired motion state. If an error occurs, the system dynamically adjusts the motor drive signal to correct the current speed or direction deviation, achieving closed-loop control based on an error compensation mechanism. Step 5: Dynamic path update and abnormal protection mechanism; 5.1 The control system regularly verifies and updates the obstacle course's current trajectory to ensure it matches the training requirements. The system's task scheduler, based on pre-defined priority logic, ensures that voice commands are processed before remote control commands, thereby ensuring the immediacy and priority of the instructor's on-site voice control. 5.2 The system also has a multi-level anomaly detection mechanism: The remote control is equipped with an emergency stop button, which corresponds to the system interruption entrance; Once the sensor data is detected to have an abnormal mutation or exceeds the set safety threshold, the system will immediately switch to a safe state and terminate the movement; If the device does not receive any valid instructions for a long time, the system will automatically enter low-power standby mode to prevent continued operation in an uncontrolled state.

[0017] In this system, coaches can remotely control moving obstacles, establish connections with athletes wearing smart belts, and follow their movements, providing defensive coverage. Only when an athlete's speed exceeds the obstacle can they escape. Laser cameras accurately measure an athlete's running speed and distance. Furthermore, the integration of AI voice control with a large AI model enables more intelligent data analysis and precise processing, providing athletes with more professional exercise plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a stereoscopic diagram of a trainer being able to remotely move an obstacle course provided by an embodiment of the present invention; Figure 2 This is a front view of a trainer being able to remotely move an obstacle course provided by an embodiment of the present invention; Figure 3This is a left view of a trainer-controlled obstacle course provided by an embodiment of the present invention; Figure 4 This is a bottom view of a trainer being able to remotely move an obstacle course provided by an embodiment of the present invention; Figure 5 This is a structural diagram of a trainer-controlled movable obstacle course provided by an embodiment of the present invention.

[0019] Figure 6 This is a diagram of the internal structure of a casing provided by an embodiment of the present invention; Figure 7 is a structural diagram of a motor provided by an embodiment of the present invention; Figure 8 This is a structural diagram of a smart belt provided by an embodiment of the present invention; In the picture: 1. Housing; 2. Universal wheel; 3. Display; 4. Charging port; 5. USB port; 6. Laser camera; 7. Controller; 8. AI voice device; 9. Speed sensor; 10. Wireless transmitter and receiver; 11. Motor; 12. Smart belt. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] like Figure 1 、 Figure 2 As shown, an embodiment of the present invention provides a trainer-controlled remote-controlled obstacle device comprising: Housing 1, universal wheel 2, laser camera 6, charging port 4, USB port 5, display 3, controller 7, AI voice device 8, speed sensor 9, wireless transmitter and receiver 10, motor 11.

[0022] like Figure 3 、 Figure 4 As shown, universal wheels 2 are fixed to both sides of the bottom of the casing 1 by screws; a laser camera 6 is embedded in the center of the front of the casing 1; a charging port 4 is provided on the left side of the right side of the casing 1; a USB port 5 is provided on the right side of the casing 1; a display 3 is fixed to the center of the top surface of the casing 1 by screws; and a controller 7 is fixed to the center of the casing 1 by screws.

[0023] An AI voice device 8 is fixed to the lower left side of the housing 1 provided in the embodiment of the present invention by screws.

[0024] A speed sensor 9 is fixed on the upper left side of the housing 1 provided in the embodiment of the present invention by screws.

[0025] A wireless transmitter and receiver 10 is fixed on the upper right side of the housing 1 provided in the embodiment of the present invention by screws.

[0026] like Figure 7 As shown, a motor 11 is fixed by screws at the lower right side of the housing 1 provided in the embodiment of the present invention.

[0027] The controller 7 provided in the embodiment of the present invention is respectively connected to the AI voice device 8, the speed sensor 9, the wireless transmitter and receiver 10, and the motor 11 through circuit lines.

[0028] like Figure 8 As shown, the embodiment of the present invention provides a coach who can remotely control the mobile obstacle device using a smart belt 12 connected to a wireless transmitter and receiver for establishing contact with the athlete to achieve the purpose of mobile obstacle following.

[0029] The smart belt 12 provided in the embodiment of the present invention can simultaneously control three or more coaches to remotely move obstacles.

[0030] An embodiment of the present invention provides a mobile obstacle course system with remote control and intelligent following capabilities. The system structurally comprises: a housing 1, universal wheels 2, a laser camera 6, a charging port 4, a USB port 5, a display 3, a controller 7, an AI voice control device 8, a speed sensor 9, a wireless transmitter and receiver 10, and a motor 11. The housing utilizes a modular design, offering excellent structural stability and spatial layout optimization capabilities, enabling efficient integration and precise positioning of various functional components. The system can adjust the movement path in real time through remote control commands and an intelligent following algorithm, providing trainers with highly accurate and diverse training assistance functions.

[0031] Universal wheels 2 are fixedly mounted on both sides of the bottom of the housing 1, ensuring the device's multi-directional motion on a flat surface. Driven by a motor 11, the universal wheels control the device's movement. Combined with a speed sensor 9, they provide precise measurement and feedback control of movement speed. In conjunction with a controller 7, closed-loop control of movement speed, direction, and acceleration is achieved, ensuring stable movement and high-precision regulation. The speed sensor is mounted on the upper left side of the housing to provide optimal signal transmission and data acquisition conditions.

[0032] A laser camera 6 is embedded in the center of the front of the housing 1, capturing depth information and visual feature data of the environment. The laser camera possesses high-precision three-dimensional perception capabilities, generating a depth map and spatial point cloud data of the environment through active laser scanning and imaging algorithms. A controller 7 processes the laser camera's data stream in real time to perform comprehensive calculations for target recognition, path planning, and obstacle avoidance strategies. This visual perception module provides the device with global environmental awareness, enhancing its intelligent navigation and dynamic tracking capabilities in complex environments.

[0033] The present invention incorporates an AI voice device 8 located on the lower left side of the housing 1 for human-computer interaction and command parsing. The AI voice device utilizes a deep learning algorithm to perform real-time recognition and parsing of the trainer's voice commands, converting the results into control commands that are then input into the controller 7. This module supports speech recognition in multiple languages and in noisy environments. Through a model architecture combining convolutional neural networks with recurrent neural networks, it achieves highly accurate speech recognition and command parsing in complex environments, providing a more intelligent interaction method for the device.

[0034] A wireless transmitter and receiver 10 is fixed to the upper right side of the housing 1 of the present invention, which is used to establish a wireless communication connection with the smart belt 12. The wireless transmitter and receiver utilizes high-frequency radio frequency communication technology, featuring long-range, low-latency, and anti-interference transmission characteristics. Through the optimized design of the data link protocol and communication protocol stack, the module can simultaneously handle the communication needs of multiple devices, achieving one-to-many collaborative control capabilities. The smart belt 12 is capable of synchronous communication with the wireless transmitter and receiver, thereby ensuring the coordinated movement and precise control of multiple mobile obstacle devices in training scenarios.

[0035] The smart belt 12 of the present invention acts as a signal source device worn by athletes. By coupling signals with the wireless transmitter and receiver 10, the device can sense and track the athlete's position and movement trajectory in real time. The smart belt can calibrate its position using beacon positioning technology or inertial navigation technology, and exchange data with the device's control system via a wireless communication module. By integrating target tracking algorithms with path optimization algorithms, the device can track athletes in real time with high precision in dynamic environments and generate highly adaptable movement paths.

[0036] The control system of the present invention, centered around the controller 7, utilizes a modular architecture integrated with an embedded processor to enable real-time data acquisition, analysis, and control command scheduling for each functional module. The control system possesses adaptive learning and parameter optimization capabilities. Through comprehensive analysis of environmental data, motion status, and user commands, it enables customized settings and automatic switching between different training modes. The controller is connected to the AI voice device, speed sensor, wireless transmitter / receiver, and motor via circuit lines to ensure stable data transmission and efficient command execution.

[0037] The intelligent mobile obstacle system of this invention is suitable for a variety of sports training and rehabilitation scenarios, demonstrating exceptional adaptability and stability in personalized and high-precision training tasks. The device uses a controller to switch between different motion modes and optimize parameters in real time, enabling the simultaneous control and independent scheduling of multiple mobile obstacle devices. The intelligent belt 12 can simultaneously coordinate control with three or more mobile obstacle devices, providing technical support and system assurance for multi-target tracking and precise movement in complex training scenarios.

[0038] An embodiment of the present invention provides a trainer-controlled remote-control system for a mobile obstacle device based on the trainer-controlled remote-control system. The system includes a central control unit, a wireless communication module, a voice recognition processor, a motor drive unit, and a multi-source sensor interface module, wherein: The central control unit is located in the middle area of the housing and is used to run the embedded control program to implement instruction parsing, state scheduling and path updating; The wireless communication module includes a transmitter and a receiver, which establishes a point-to-point connection with an external remote control terminal based on WiFi or Bluetooth communication protocol, and is used to receive control instructions issued by the coach and transmit them to the central control unit; The speech recognition processor is connected to the AI speech device, supports the local semantic parsing algorithm, and after recognizing the keywords, forwards the control intention to the central control unit in the intermediate instruction format; The motor drive unit is linked to the central control unit to adjust the motor speed through the PWM signal to achieve closed-loop control of the obstacle movement state; The multi-source sensor interface module is connected to the speed sensor, position encoder and attitude sensor to build a real-time state feedback path for the device and support adaptive correction of motion parameters based on the PID algorithm.

[0039] The central control unit adopts a microcontroller architecture based on the STM32 or ARM CortexM series, is embedded with a real-time operating system, and concurrently manages voice control, remote control response, and speed closed-loop adjustment tasks using a task priority queue scheduling strategy.

[0040] The speech recognition processor integrates an end-side inference engine, uses a lightweight neural network model for keyword recognition, and performs asynchronous data exchange with the central control unit via serial communication or SPI bus; 1) The voice signal data processing process begins with the AI voice device collecting sound waves from the coach's commands. A MEMS microphone array is used for beamforming to improve the signal-to-noise ratio of the commands. The voice signal then enters the voice recognition processor for on-device inference. This processor, with a built-in lightweight neural network model, performs keyword classification and recognition through Mel-spectrogram extraction and feature vector quantization. The parsed control intent is encapsulated into a structured intermediate control frame and sent to the central control unit via the SPI or UART interface. The task scheduler then loads it into the command buffer queue for subsequent parsing and execution. 2) The remote control wireless command link establishes a low-latency point-to-point communication channel based on the WiFi Direct or BLE protocol stack. Control commands sent by the instructor's remote control are encapsulated in simplified control frames at the data link layer and sent to the system's wireless communication module via the GATT or UDP protocol. Upon receiving the frame data, the receiver performs header check, CRC check, and channel error correction, and then passes the payload to the central control unit for unpacking and parsing. Command content, including target speed, direction angle, or stop flag, is encoded in a standard fixed-length byte format to ensure real-time and stability. 3) Control command parsing and motor control signal generation are triggered by the central control unit's RTOS-based event-driven mechanism. The parsed control commands are sent via an internal bus to the motor drive unit, which converts the target speed into a duty cycle adjustment value and outputs a PWM signal to drive a brushed or brushless DC motor. The drive unit has a built-in H-bridge control circuit and current protection module to ensure output response time within milliseconds, improving overall machine response sensitivity. In addition, the system supports dynamic speed limiting and steering adjustment logic, dynamically setting the motor's target operating range through limit registers, enhancing control stability and safety margin control capabilities. 4) The multi-source state feedback processing mechanism is completed by the sensor fusion subsystem formed by the speed sensor, encoder, and attitude sensor. The speed sensor provides linear velocity estimation, the rotary encoder outputs pulse counts to convert angular displacement, and the attitude sensor outputs three-dimensional attitude angles through a combination of six-axis or nine-axis gyroscopes and accelerometers. All sensor data is uniformly sampled by the ADC or I2C / SPI transmission channel, and the state fusion algorithm built into the central control unit performs data integration. The final system uses an incremental PID regulator to compensate for the current error and dynamically adjust the PWM signal to achieve real-time closed-loop correction of the motion state and adaptive adjustment of the path deviation.

[0041] An embodiment of the present invention provides a control method based on the above-mentioned trainer-controlled remote control system for a mobile obstacle device, the method comprising: Step 1: Voice command collection and edge semantic processing; 1.1 The coach's voice commands are first collected by an AI voice device. The device's front-end uses a pickup system consisting of a micro-microphone array. This system uses a spatial filtering algorithm to enhance the forward voice signal while reducing ambient noise interference. 1.2 The pre-processed speech signal is fed into the speech recognition processor, where edge inference is performed using a locally deployed lightweight neural network model. The model first converts the raw audio signal into a Mel-spectrogram, extracts vectors representing semantic features, and finally inputs these vectors into the neural network for keyword classification and recognition. 1.3 Once a target keyword is identified with high confidence, the speech recognition processor converts the recognition result into an intermediate control frame and transmits it asynchronously to the central control unit via serial communication or SPI bus; Step 2: Remote control command link establishment and data analysis; 2.1 The instructor's handheld remote control terminal establishes a low-latency, point-to-point data channel with the device's wireless communication module via WiFi Direct or BLE low-power protocols. This channel features broadcast discovery, fast connection, and handshake confirmation mechanisms. 2.2 The control command is encapsulated into a concise data frame through the protocol, which includes the command type, target value, command timestamp and redundant bits for verification; 2.3 After the receiving wireless module completes the structural analysis and verification of the data frame, it unpacks the command content and forwards it to the command processing module of the central control unit, which adds it to the control command queue for execution; Step 3: Central control logic analysis and motor drive signal generation; 3.1 The central control unit is based on an embedded real-time operating system and uses a task scheduler to manage all input events, including voice commands, remote control input, and sensor feedback; 3.2 Whenever a new control command is added to the buffer queue, the system initiates the motor control task based on the command type, converting the received target speed and direction into a motor speed control signal. During this process, the central control unit calculates the PWM signal output parameters corresponding to the target motion state and sends them to the motor drive unit. 3.3 The motor drive unit includes a complete power regulation circuit and current protection mechanism, which can effectively prevent system instability caused by sudden load changes while ensuring responsiveness. In addition, dynamic speed and steering limit logic is provided to dynamically set the operating range based on the training environment or mission requirements, improving the control safety of the equipment. Step 4: Multi-source sensor feedback and closed-loop motion control; 4.1 The control system uses a sensor fusion path formed by speed sensors, encoders, and attitude sensors to perform high-frequency sampling and feedback on the real-time status of the equipment; The speed sensor is used to measure the actual operating speed of the equipment; Encoders detect changes in displacement or angle; The attitude sensor provides three-dimensional attitude information of the device, including parameters such as roll, pitch, and yaw; 4.2 All sensor data is uniformly transmitted to the control unit via the system's internal I²C or SPI interface, and is integrated using a fusion algorithm to filter out noise and fill in missing information; 4.3 The control unit determines the current motion state based on the fused data and compares it with the desired motion state. If an error occurs, the system dynamically adjusts the motor drive signal to correct the current speed or direction deviation, achieving closed-loop control based on an error compensation mechanism. Step 5: Dynamic path update and abnormal protection mechanism; 5.1 The control system regularly verifies and updates the obstacle course's current trajectory to ensure it matches the training requirements. The system's task scheduler, based on pre-defined priority logic, ensures that voice commands are processed before remote control commands, thereby ensuring the immediacy and priority of the instructor's on-site voice control. 5.2 The system also has a multi-level anomaly detection mechanism: The remote control is equipped with an emergency stop button, which corresponds to the system interruption entrance; Once the sensor data is detected to have an abnormal mutation or exceeds the set safety threshold, the system will immediately switch to a safe state and terminate the movement; If the device does not receive any valid instructions for a long time, the system will automatically enter low-power standby mode to prevent continued operation in an uncontrolled state.

[0042] Example 1: Application of the Intelligent Mobile Obstacle System in Football Training In soccer training scenarios, the intelligent mobile obstacle system of the present invention simulates dynamic defensive or offensive obstacles to enhance players' tactical awareness and flexibility. Universal wheels 2 are mounted on the bottom of the training device's housing 1, enabling full-scale movement on grass or smooth surfaces. The coach wears a smart belt 12, which communicates with multiple mobile obstacles via a wireless transmitter-receiver 10.

[0043] The laser camera 6 configured in the system performs real-time three-dimensional scanning of the training area, generates point cloud data and constructs a dynamic scene model. Through the AI voice device 8, the coach can input training goals in natural language instructions, such as "simulate high-pressure defense" or "maintain distance between defense lines." The controller 7 parses the voice instructions and converts them into control commands, and dynamically plans the motion path in combination with the data collected by the laser camera. During training, the mobile obstacle can adjust its position and movement speed according to the coach's instructions, and achieve precise movement by driving the universal wheel through the motor 11. At the same time, the speed sensor 9 monitors the movement status in real time to ensure that the device executes according to the predetermined path and speed. The system supports multi-device collaborative control, and the smart belt 12 can control three or more mobile obstacles at the same time to achieve multi-target tracking and control in complex training scenarios.

[0044] Through the optimized design of the wireless communication module, the system achieves low-latency, interference-resistant data transmission within a 50-meter range. During actual training, the device maintains a stable position using a ratchet-pin self-locking mechanism, providing coaches with more flexible and diverse training options. This example demonstrates the system's significant application effectiveness and high-precision control capabilities in soccer training.

[0045] Example 2: Application of the Intelligent Mobile Obstacle System in Rehabilitation Training In rehabilitation training scenarios, the intelligent mobility obstacle system of the present invention is used to assist patients in lower limb rehabilitation training and gait correction, improving their motor control and recovery by providing mobility obstacles and dynamic guidance. The device's housing 1 adopts a lightweight design to reduce the effects of load and inertia during exercise.

[0046] The patient wears a smart belt 12, which communicates with the mobility barrier system in real time via a wireless transmitter-receiver 10. A laser camera 6 within the device collects information about the patient's movement and surrounding environment, generating three-dimensional motion trajectories and gait analysis data. The system's AI voice control 8 receives voice commands from the patient or rehabilitation therapist and automatically adjusts training difficulty and path planning.

[0047] During rehabilitation training, controller 7 combines data collected by the laser camera with information fed back by speed sensor 9 to calculate the patient's movement trajectory and the path of obstacles in real time. Motor 11 drives universal wheels 2 to move along the ground at a set speed and trajectory, providing patients with movement guidance and training challenges around dynamic obstacles.

[0048] The Energy Belt 12 features motion status monitoring and posture assessment, collecting the patient's gait characteristics and posture changes through an inertial measurement unit (IMU). The system can adaptively adjust the training path and speed based on the patient's real-time status, and achieve coordinated operation and intelligent scheduling of multiple mobile obstacle devices through wireless communication.

[0049] Through adaptive optimization and adjustment of control system parameters, the system can be flexibly applied in different rehabilitation training scenarios and provide patients with personalized and dynamic rehabilitation training plans. Actual application results have shown that the system can significantly improve patients' motor control accuracy and rehabilitation efficiency.

[0050] These two examples demonstrate the specific implementation methods and technical effects of the intelligent mobile obstacle system in different application scenarios. The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited to this. Any modifications, equivalent substitutions, and improvements made by any person skilled in the art within the technical scope disclosed by the present invention, within the spirit and principles of the present invention, shall be covered by the scope of protection of the present invention.

Claims

1. A trainer can remotely control a mobile obstacle course, characterized in that: The coach can remotely move the obstacle device including: Casing, universal wheel, laser camera, charging port, USB port, display, controller, AI voice device, speed sensor, wireless transmitter and receiver, motor; Universal wheels are fixed on both sides of the bottom of the casing by screws; a laser camera is embedded in the center of the front of the casing; a charging port is opened on the left right side of the casing; a USB port is opened on the right right side of the casing; a display is fixed in the center of the top of the casing by screws; and a controller is fixed in the center of the casing by screws.

2. The trainer can remotely control the obstacle course according to claim 1, wherein: An AI voice device is fixed by screws on the lower left side of the casing.

3. The trainer can remotely control the obstacle course according to claim 1, wherein: A speed sensor is fixed on the upper left side of the casing by screws.

4. The trainer can remotely control the obstacle course according to claim 1, wherein: A wireless transmitter and receiver is fixed on the upper right side of the casing by screws.

5. The trainer can remotely control the obstacle course according to claim 1, wherein: An electric motor is fixed to the lower right side of the casing by screws; the controller is connected to the AI voice device, speed sensor, wireless transmitter and receiver, and electric motor through circuit lines.

6. The trainer can remotely control the obstacle course according to claim 1, wherein: The coach can remotely control the mobile obstacle course using a smart belt that is connected to a wireless transmitter and receiver to establish contact with the athlete so that the obstacle course can follow the athlete.

7. The trainer can remotely control the obstacle course according to claim 9, wherein: The smart belt can control three or more coaches to move the obstacle device remotely at the same time.

8. A trainer-controlled remote control system for a movable obstacle device based on the trainer-controlled remote control movable obstacle device according to claim 1, characterized in that: The system includes a central control unit, a wireless communication module, a speech recognition processor, a motor drive unit, and a multi-source sensor interface module, wherein: The central control unit is located in the middle area of the housing and is used to run the embedded control program to implement instruction parsing, state scheduling and path updating; The wireless communication module includes a transmitter and a receiver, which establishes a point-to-point connection with an external remote control terminal based on WiFi or Bluetooth communication protocol, and is used to receive control instructions issued by the coach and transmit them to the central control unit; The speech recognition processor is connected to the AI speech device, supports the local semantic parsing algorithm, and after recognizing the keywords, forwards the control intention to the central control unit in the intermediate instruction format; The motor drive unit is linked to the central control unit to adjust the motor speed through the PWM signal to achieve closed-loop control of the obstacle movement state; The multi-source sensor interface module is connected to the speed sensor, position encoder and attitude sensor to build a real-time state feedback path for the device and support adaptive correction of motion parameters based on the PID algorithm.

9. The trainer-controlled remote control system for a mobile obstacle course according to claim 8, wherein: The central control unit adopts a microcontroller architecture based on the STM32 or ARM CortexM series, is embedded with a real-time operating system, and concurrently manages voice control, remote control response, and speed closed-loop adjustment tasks using a task priority queue scheduling strategy.

Images