Intelligent wheelchair automatic navigation system and method

Abstract

The invention belongs to wheelchairs, and particularly relates to an automatic navigation system and method for an intelligent wheelchair, and the method comprises the following steps: 1, after the system is started, a visual sensor assembly module starts to work to monitor the posture and position information of a user, and a camera captures the body posture of the user and the relative position information of the wheelchair; the data is transmitted to the control and execution module for analysis and processing; 2, the user can send a calling signal through a voice command or a mobile phone application, and the calling signal informs the system of the demand of the user; 3, after the motion control and wheelchair navigation module receives the signal of the user and the sensor data, an optimal navigation route is formulated according to the requirements of the user; 4, a navigation route is determined, a wheelchair body starts a motor according to an instruction of a motion control and wheelchair navigation module, and the wheelchair can avoid obstacles according to an environment map; 5, the wheelchair arrives at the target position, the user conveniently gets on and gets off the vehicle, and the system can automatically stop the vehicle according to instructions or gestures of the user.

Description

Technical Field

[0001] The present invention relates to a wheelchair, and in particular to an intelligent wheelchair automatic navigation system and method. Background Art

[0002] When a traditional wheelchair is not in use, it needs to be moved to a specific parking position by a family member or staff. When the user needs to use it again, someone else is still needed to move the wheelchair from the parking position to the user's side before it can be used. This process cannot be completed by the user himself and requires the participation of other people, which brings inconvenience to the user. Therefore, a smart wheelchair system that can dynamically monitor user needs and navigate autonomously is needed.

[0003] In view of the above technical defects, a solution of an intelligent wheelchair automatic navigation system and method is now proposed. Summary of the invention

[0004] To solve the above problems, the present invention provides the following technical solutions: An intelligent wheelchair automatic navigation system and method, comprising a visual sensor module, a control and execution module, a motion control and wheelchair navigation module, a user feedback and system optimization module, a trajectory analysis and preference learning module and a user interface module; The visual sensor component module includes but is not limited to a high-resolution camera, a laser radar, and a depth sensor, which are responsible for real-time monitoring of the environment around the wheelchair, capturing the user's posture information and position data, and transmitting the captured data information to the control execution module; The control and execution module includes a data processing and fusion unit and a path planning and navigation unit. The control and execution module receives sensor data information from the visual sensor component module, processes the information through the data processing and fusion unit, and transmits the processed information to the path planning and navigation unit. The path planning and navigation unit determines the best path for the wheelchair based on user needs, environmental information, and historical trajectory analysis and preferences, and transmits the path information to the motion control and wheelchair navigation module; The motion control and wheelchair navigation module includes a PID controller, a wheelchair body and an obstacle avoidance and dynamic adjustment unit. The wheelchair body is the physical part of the intelligent wheelchair, including an electric wheelchair chassis, a seat and a battery. These components work together and based on the path planning data of the path planning and navigation unit in the control and execution module, the wheelchair can control the wheelchair body to move along the path of the control and execution module according to the instructions of the PID controller. During the movement of the wheelchair body, the path route is dynamically adjusted by the obstacle avoidance and dynamic adjustment unit, and the data information is transmitted to the user feedback and system optimization module in real time; The user feedback and system optimization module, the system will collect real-time feedback from users through voice recognition, touch screen, buttons, etc., and upload the feedback information to the motion control and wheelchair navigation module in real time. The motion control and wheelchair navigation module will process the received information and combine it with the real-time trajectory analysis and preference learning module to control the route re-planning and design until the wheelchair journey is completed.

[0005] The historical trajectory analysis and preference learning module is based on the final data of the motion control and wheelchair navigation modules. The PDI controller transmits the data information of the path to the historical trajectory analysis and preference learning module. The system will regularly collect the user's historical travel trajectory data and use machine learning algorithms to identify the user's travel mode. The system can predict the user's future travel needs and recommend a route that is more in line with their habits. The user interface module is a way for users to interact with the system, usually including voice recognition, buttons, and mobile phone applications. Users can use the user interface module to control the movement status of the wheelchair in real time, view the walking trajectory route, and timely feedback and save data information such as the historical trajectory analysis and preference learning module, user feedback and system optimization module.

[0006] Further, the visual sensor component module includes a high-resolution camera, a laser radar and a depth sensor; The camera is used to capture the user's visual information, including facial expressions, gestures, head posture, etc., so that the system can better understand the user's needs; The laser radar measures the distance and shape of the surrounding environment through the reflection of laser beams to build maps and detect obstacles; The sensor is used to obtain the distance and relative position between the user and the wheelchair to ensure safe boarding and alighting; Furthermore, the control and execution module includes a data processing and fusion unit and a path planning and navigation unit. The data processing and fusion unit performs data processing and analysis as follows: The data processing and fusion unit data includes user position coordinate data processing and user sitting angle data processing: the user position coordinate data, that is, the environment image of the user and the wheelchair captured by the depth camera or laser radar, the user position coordinate is in the two-dimensional control key, and the user coordinate is recorded as , represents the coordinates of the user on the ground; the posture angle data can be obtained through sensor data, recorded as , the coordinate system of the wheelchair is The relative position of the user and the wheelchair can be calculated by rotation and translation. The rotation matrix Used to indicate changes in attitude angle: if is the coordinate of the user relative to the wheelchair (in the wheelchair coordinate system), then .

[0007] Furthermore, the data processing and fusion unit also includes a Kalman filter algorithm. The data processing and fusion unit processes the user's position coordinate data based on the dynamic environment and the changing posture and the user's sitting angle data, and uses algorithms such as Kalman filtering to continuously update and optimize the user's position and posture. The Kalman filter formula is as follows: in is the estimated state at the current moment, represents the observed sensor data, It represents the observation matrix; represents the Kalman gain; The algorithm can dynamically adjust the user's position estimate based on sensor data and system models to improve positioning accuracy.

[0008] Furthermore, the path planning and navigation unit calculates the best path according to the user's current position and target position based on the data information sent by the control and execution module and the historical trajectory information of the historical trajectory analysis and preference learning module. During the planning process, the system will continuously optimize the path, avoid obstacles and congested areas in real time; the path planning and navigation unit includes environment modeling and obstacle detection, path planning algorithm and path optimization. The environment modeling and obstacle detection are performed by laser radar, ultrasonic sensor or camera. The system can detect surrounding obstacles and record the obstacle set as: Each obstacle Each has its place The influence of obstacles will be avoided during the path planning process. For example, when calculating the path, the path that avoids obstacles is the valid path; The path planning algorithm is based on obstacle data and data information from the data processing and fusion unit. Algorithm, The algorithm uses heuristic search to find the starting point To the destination The shortest path of The cost function of the algorithm is: in represents the actual cost from the starting point to the current node n, is the heuristic estimated cost from the current node n to the target point, usually using the Euclidean distance: The path optimization, when the obstacle changes or the user's posture changes, the path needs to be dynamically adjusted, and the dynamic programming or online optimization algorithm is used to adjust the path as follows: in are the adjacent nodes on the path, is the Euclidean distance between two points.

[0009] Furthermore, the motion control and wheelchair navigation module includes a PID controller, a wheelchair body and an obstacle avoidance and dynamic adjustment unit. The PID controller is based on the data controlled by the control and execution module. The current position of the wheelchair is , the target location is The error is: Output control signal of PID controller for: in is the proportionality factor, used to adjust the error, is the integration factor, used to eliminate long-term errors is the differential coefficient, used to predict the error change; Through the PID controller, the system can dynamically adjust the speed and direction of the wheelchair according to the real-time error to ensure that the wheelchair moves along the planned path.

[0010] Furthermore, the obstacle avoidance and dynamic adjustment, during the movement of the wheelchair, obstacles may suddenly appear. In order to avoid collision, the wheelchair needs to monitor the surrounding environment in real time and adjust the path. The obstacle avoidance and dynamic adjustment include a dynamic window method, which includes a wheelchair body state space, a wheelchair body speed window, obstacle detection and evaluation, an evaluation function, and selection of the optimal speed. The dynamic window method calculates the current available speed window of the wheelchair body in real time, and then selects a suitable speed combination to avoid obstacles, and plans the optimal motion trajectory on this basis. The obstacle avoidance and dynamic adjustment are specifically analyzed as follows: The wheelchair main body state space, the current speed of the wheelchair main body is ,in is the linear velocity, is the angular velocity; The wheelchair body speed window is based on the data of the wheelchair body state space according to the maximum linear speed of the wheelchair body. and maximum angular velocity , the dynamic window defines the range of speeds that can be selected, and the boundaries of the window are usually: Line speed: Angular velocity: The obstacle detection and assessment uses sensors (such as lidar, ultrasonic sensors, etc.) to detect the distance of surrounding obstacles in real time, and calculates the collision risk of each speed combination based on the position of the obstacles; The evaluation function is for each candidate speed combination , calculate the following function To measure its quality: in Evaluate whether the current velocity combination helps to move toward the target point, usually as a function of the distance or direction to the target point; Evaluate whether the speed combination avoids the obstacle, usually based on the shortest distance to the obstacle; Consider the dynamic constraints of the wheelchair body to ensure that the speed combination does not exceed the dynamic limit of the wheelchair body. is the weight coefficient used to balance various evaluation factors; The optimal speed is selected for all candidate speed combinations. Select the evaluation value The maximum speed is taken as the final selected speed and used as the control input of the wheelchair.

[0011] Furthermore, the historical trajectory analysis and preference learning module is based on the final data of the motion control and wheelchair navigation modules. The PDI controller transmits the data information of the path to the historical trajectory analysis and preference learning module. The system will regularly collect the user's historical travel trajectory data and use machine learning algorithms to identify the user's travel mode. The system can predict the user's future travel needs and recommend a route that is more in line with their habits. The user interface module, the user interface is the way for users to interact with the system, usually including voice recognition, buttons, mobile phone applications, etc. Users can interact with the wheelchair in the following ways: The buttons and touch screen: The buttons or touch screen on the wheelchair allow users to send call signals, start navigation and provide feedback; The voice command: The user can tell the wheelchair their needs through voice commands, such as "forward", "stop", etc. The mobile phone application: The wheelchair can be connected to the user's smartphone application, and the user can monitor the wheelchair position and status in real time through the application, and send instructions.

[0012] Furthermore, according to the method based on an intelligent wheelchair automatic navigation system according to claim 9, it is characterized in that the intelligent wheelchair automatic navigation system according to any one of claims 1 to 9 is adopted, and the steps are as follows: Step 1: First, after the system is started, the visual sensor component module starts working to monitor the user's posture and position information. The camera captures the user's body posture and relative position information with the wheelchair. At the same time, the laser radar and depth sensor detect the distance and relative position between the user and the wheelchair. These data are transmitted to the control and execution module for analysis and processing; Step 2: The user can send a summon signal through voice commands or mobile phone applications, which informs the system of the user's needs, such as needing to get on or off the vehicle, or move to a certain location. Step 3: After receiving the user's signals and sensor data, the motion control and wheelchair navigation module will develop the best navigation route based on the user's needs and environmental conditions. Taking into account obstacles in the environment, the navigation algorithm will determine a safe path to ensure that the wheelchair can reach the target location smoothly; Step 4: Once the navigation route is determined, the wheelchair body will start the motor according to the instructions of the motion control and wheelchair navigation module and begin autonomous navigation. The wheelchair will avoid obstacles according to the environmental map and adjust the speed and direction to ensure safe navigation; Step 5: When the wheelchair reaches the target location, the user can get on and off easily. The system can automatically stop the vehicle according to the user's instructions or gestures, and help the user get on and off as needed.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. In the intelligent wheelchair automatic navigation system and method of the present invention, the present invention adopts The algorithm's path planning method can accurately calculate the shortest path from the starting position to the target position, avoiding the problems of lengthy paths or inability to avoid obstacles that may occur in traditional navigation systems. By combining real-time environmental perception data, the wheelchair can automatically select the best path in complex indoor environments to ensure that it reaches the destination quickly and safely. 2. In the automatic navigation system and method of an intelligent wheelchair of the present invention, the present invention uses a laser radar and a depth sensor to scan the surrounding environment in real time, and can accurately detect obstacles in front of and around the wheelchair, and can take avoidance measures in time. Compared with the traditional static obstacle detection method, the system can adjust the path in real time according to environmental changes, thereby avoiding collisions and falling accidents, and improving the safety and reliability of the wheelchair; 3. In the automatic navigation system and method of an intelligent wheelchair of the present invention, the PID controller can adjust the speed and steering angle of the wheelchair in real time according to the error between the wheelchair and the target position, so as to ensure that the wheelchair can travel smoothly and accurately in a complex environment. This not only improves the comfort of the user, but also can maintain good driving performance under different ground conditions, avoiding the instability problem that may occur in traditional wheelchairs when turning or accelerating; 4. The intelligent wheelchair automatic navigation system and method of the present invention can not only realize automatic navigation, but also assist users in getting on and off the vehicle. Through the coordinated work of depth perception and visual sensors, the system can determine whether the user has completed the operation of getting on and off the vehicle safely, and adjust the position of the wheelchair as needed, thereby avoiding falls or injuries caused by improper operation. This intelligent assistance function greatly improves the quality of life and autonomy of the elderly, the disabled and other people with limited mobility; 5. In the intelligent wheelchair automatic navigation system and method of the present invention, the system can adjust the navigation strategy in real time according to the changes in the environment and the needs of the user. For example, when encountering sudden obstacles or narrow passages, the system can re-plan the path and quickly avoid the obstacles to ensure the safety of the user. This highly intelligent processing capability enables the wheelchair to show good adaptability and autonomy in a dynamic environment, avoiding the limitations of traditional manually controlled wheelchairs in complex environments; 6. The intelligent wheelchair automatic navigation system and method of the present invention integrates multiple sensors (such as lidar and depth camera) through modular design, making the hardware structure of the system more concise and compact, easy to maintain and upgrade, and easy for users to operate. The automatic navigation system can be started through simple voice commands or manual input of the target location, avoiding cumbersome operating steps, allowing users to use the intelligent wheelchair more easily. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings; Figure 1 A schematic diagram of the overall steps of an intelligent wheelchair automatic navigation system and method of the present invention; Figure 2 This is a schematic diagram of the overall structure of an intelligent wheelchair automatic navigation system and method of the present invention; Figure 3 The control and execution modules of the intelligent wheelchair automatic navigation system and method of the present invention include a diagram; Figure 4 The present invention discloses an intelligent wheelchair automatic navigation system and method, including motion control and wheelchair navigation modules. DETAILED DESCRIPTION

[0015] 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.

[0016] like Figure 1-Figure 4 As shown, an intelligent wheelchair automatic navigation system and method include a visual sensor module, a control and execution module, a motion control and wheelchair navigation module, a user feedback and system optimization module, a trajectory analysis and preference learning module and a user interface module; Visual sensor component modules, including but not limited to high-resolution cameras, laser radars, and depth sensors, which are responsible for real-time monitoring of the environment around the wheelchair, capturing the user's posture information and position data, and transmitting the captured data information to the control execution module; The system uses lidar, depth sensors and visual sensors to scan the surrounding environment in real time, collect information on the location, size and dynamic changes of surrounding obstacles, and through sensor data fusion, the system can build an accurate environmental model and update it in real time.

[0017] The control and execution module includes a data processing and fusion unit and a path planning and navigation unit. The control and execution module receives sensor data information from the visual sensor component module, processes the information through the data processing and fusion unit, and transmits the processed information to the path planning and navigation unit. The path planning and navigation unit determines the best path for the wheelchair based on user needs, environmental information, and historical trajectory analysis and preferences, and transmits the path information to the motion control and wheelchair navigation module; Path planning: Based on the target location and environmental perception data input by the user, the system calculates the shortest and safest driving path. If encountering sudden obstacles, the system will adjust the path in real time to avoid the obstacles and keep the target location unchanged.

[0018] The motion control and wheelchair navigation module includes a PID controller, a wheelchair body and an obstacle avoidance and dynamic adjustment unit. The wheelchair body is the physical part of the intelligent wheelchair, including an electric wheelchair chassis, a seat and a battery. These components work together and based on the path planning data of the path planning and navigation unit in the control and execution module, the wheelchair can control the wheelchair body to move along the path of the control and execution module according to the instructions of the PID controller. During the movement of the wheelchair body, the path route is dynamically adjusted through the obstacle avoidance and dynamic adjustment unit, and the data information is transmitted to the user feedback and system optimization module in real time; Based on the path planning results, the system adjusts the speed and direction of the electric drive system through the PID control algorithm to accurately control the movement of the wheelchair. The PID controller dynamically adjusts the driving direction and speed based on the error between the wheelchair and the target position, the actual deviation during the movement, and other information to ensure that the wheelchair can complete navigation smoothly and accurately.

[0019] When an obstacle is detected in front or on the side of the wheelchair, the lidar and depth sensor will feedback the location of the obstacle. The system will determine the size and type of the obstacle. If it is a static obstacle, the system will calculate a detour path; if it is a dynamic obstacle (such as a pedestrian or other object), the system will adjust the speed and direction in real time to avoid collision.

[0020] User feedback and system optimization module, the system will collect real-time feedback from users through voice recognition, touch screen, buttons, etc., and upload the feedback information to the motion control and wheelchair navigation module in real time. The motion control and wheelchair navigation module will process the received information and combine it with the real-time trajectory analysis and preference learning module to control the route re-planning and design until the wheelchair journey is completed.

[0021] The historical trajectory analysis and preference learning module uses the final data of the motion control and wheelchair navigation modules to transmit the data information of the path to the historical trajectory analysis and preference learning module. The system will regularly collect the user's historical travel trajectory data and use machine learning algorithms to identify the user's travel mode. The system can predict the user's future travel needs and recommend a route that is more in line with their habits. User interface module: The user interface is the way for users to interact with the system, usually including voice recognition, buttons, and mobile phone applications. Users can use the user interface module to control the movement status of the wheelchair in real time, view the walking trajectory route, and promptly feedback and save data information such as the historical trajectory analysis and preference learning module, user feedback, and system optimization module.

[0022] Specifically, the visual sensor component module includes a high-resolution camera, a lidar, and a depth sensor; Camera: used to capture the user's visual information, including facial expressions, gestures, head posture, etc., so that the system can better understand the user's needs; LiDAR: measures the distance and shape of the surrounding environment through the reflection of laser beams to build maps and detect obstacles; Degree sensor: used to obtain the distance and relative position between the user and the wheelchair to ensure safe getting on and off; The wheelchair uses lidar to scan obstacles in front of it and uses depth sensors to obtain ground information. When the user enters the destination, the system uses The algorithm plans the shortest path and controls the wheelchair along the path through a PID controller. When encountering obstacles (such as furniture), the system will adjust the path in real time, bypass the obstacle, and continue to drive to the target location.

[0023] Specifically, the control and execution module includes a data processing and fusion unit and a path planning and navigation unit. The data processing and fusion unit performs data processing and analysis as follows: Data processing and fusion unit data includes user position coordinate data processing and user sitting angle data processing: User position coordinate data, that is, the depth camera or lidar captures the environmental image of the user and the wheelchair, the user position coordinate is in the two-dimensional control key, and the user coordinate is recorded as , represents the coordinates of the user on the ground; the posture angle data can be obtained through sensor data and is recorded as , the coordinate system of the wheelchair is The relative position of the user and the wheelchair can be calculated by rotation and translation. The rotation matrix Used to indicate changes in attitude angle: if is the coordinate of the user relative to the wheelchair (in the wheelchair coordinate system), then .

[0024] Specifically, the data processing and fusion unit also includes a Kalman filter algorithm. The data processing and fusion unit processes the user's position coordinate data based on the dynamic environment and the changing posture and user's sitting angle data, and uses algorithms such as Kalman filtering to continuously update and optimize the user's position and posture. The Kalman filter formula is as follows: in is the estimated state at the current moment, represents the observed sensor data, It represents the observation matrix; represents the Kalman gain; The algorithm can dynamically adjust the user's position estimate based on sensor data and system models to improve positioning accuracy.

[0025] Specifically, the path planning and navigation unit calculates the best path according to the user's current position and target position based on the data information sent by the control and execution module and the historical trajectory information of the historical trajectory analysis and preference learning module. During the planning process, the system will continuously optimize the path, avoid obstacles and congested areas in real time; the path planning and navigation unit includes environment modeling and obstacle detection, path planning algorithm and path optimization. Environment modeling and obstacle detection use lidar, ultrasonic sensor or camera. The system can detect surrounding obstacles and record the obstacle set as: Each obstacle Each has its place The influence of obstacles will be avoided during the path planning process. For example, when calculating the path, the path that avoids obstacles is the valid path; The path planning algorithm is based on obstacle data and data information from the data processing and fusion unit. algorithm, The algorithm uses heuristic search to find the starting point To the destination The shortest path of The cost function of the algorithm is: in represents the actual cost from the starting point to the current node n, is the heuristic estimated cost from the current node n to the target point, usually using the Euclidean distance: Path optimization: When obstacles change or user posture changes, the path needs to be adjusted dynamically. Dynamic programming or online optimization algorithms are used to adjust the path as follows: in are the adjacent nodes on the path, is the Euclidean distance between two points.

[0026] Specifically, the motion control and wheelchair navigation module includes a PID controller, a wheelchair body, and an obstacle avoidance and dynamic adjustment unit. The PID controller, based on the data controlled by the control and execution module, the current position of the wheelchair is , the target location is The error is: Output control signal of PID controller for: in is the proportionality factor, used to adjust the error, is the integration factor, used to eliminate long-term errors , is the differential coefficient, used to predict the error change; Through the PID controller, the system can dynamically adjust the speed and direction of the wheelchair according to the real-time error to ensure that the wheelchair moves along the planned path.

[0027] Specifically, obstacle avoidance and dynamic adjustment. During the movement of the wheelchair, obstacles may suddenly appear. In order to avoid collisions, the wheelchair needs to monitor the surrounding environment in real time and adjust the path. Obstacle avoidance and dynamic adjustment include the dynamic window method, which includes the wheelchair body state space, the wheelchair body speed window, obstacle detection and evaluation, evaluation function and selection of the optimal speed. The dynamic window method calculates the current available speed window of the wheelchair body in real time, and then selects the appropriate speed combination to avoid obstacles, and plans the optimal motion trajectory on this basis. The specific analysis of obstacle avoidance and dynamic adjustment is as follows: The state space of the wheelchair body, the current speed of the wheelchair body is ,in is the linear velocity, is the angular velocity; The wheelchair body speed window is based on the data of the wheelchair body state space and the maximum linear speed of the wheelchair body. and maximum angular velocity , the dynamic window defines the range of speeds that can be selected, and the boundaries of the window are usually: Line speed: Angular velocity: Obstacle detection and assessment, using sensors (such as lidar, ultrasonic sensors, etc.) to detect the distance of surrounding obstacles in real time and calculate the collision risk of each speed combination based on the position of the obstacles; The evaluation function for each candidate speed combination , calculate the following function Measure its quality: in Evaluate whether the current velocity combination helps to move toward the target point, usually as a function of the distance or direction to the target point; Evaluate whether the speed combination avoids the obstacle, usually based on the shortest distance to the obstacle; Consider the dynamic constraints of the wheelchair body to ensure that the speed combination does not exceed the dynamic limit of the wheelchair body. Weight coefficient, used to balance various evaluation factors; Select the optimal speed for all candidate speed combinations Select the evaluation value The maximum speed is taken as the final selected speed and used as the control input of the wheelchair.

[0028] Specifically, the historical trajectory analysis and preference learning module is based on the final data of the motion control and wheelchair navigation modules. The PDI controller transmits the data information of the path to the historical trajectory analysis and preference learning module. The system will regularly collect the user's historical travel trajectory data and use machine learning algorithms to identify the user's travel mode. The system can predict the user's future travel needs and recommend a route that is more in line with their habits. User interface module. The user interface is the way for users to interact with the system, usually including voice recognition, buttons, mobile applications, etc. Users can interact with the wheelchair in the following ways: Buttons and touch screens: Buttons or touch screens on the wheelchair allow users to send summon signals, start navigation and provide feedback; Voice commands: Users can tell the wheelchair their needs through voice commands, such as "go", "stop", etc. Mobile applications: The wheelchair can be connected to the user's smartphone application, and the user can use the application to monitor the wheelchair's position and status in real time and send instructions.

[0029] Specifically, according to the method based on an intelligent wheelchair automatic navigation system according to claim 9, it is characterized in that an intelligent wheelchair automatic navigation system according to any one of claims 1 to 9 is adopted, and the steps are as follows: Step 1: First, after the system is started, the visual sensor component module starts working to monitor the user's posture and position information. The camera captures the user's body posture and relative position information with the wheelchair. At the same time, the laser radar and depth sensor detect the distance and relative position between the user and the wheelchair. These data are transmitted to the control and execution module for analysis and processing; Step 2: The user can send a summon signal through voice commands or mobile phone applications, which informs the system of the user's needs, such as needing to get on or off the vehicle, or move to a certain location. Step 3: After receiving the user's signals and sensor data, the motion control and wheelchair navigation module will develop the best navigation route based on the user's needs and environmental conditions. Taking into account obstacles in the environment, the navigation algorithm will determine a safe path to ensure that the wheelchair can reach the target location smoothly; Step 4: Once the navigation route is determined, the wheelchair body will start the motor according to the instructions of the motion control and wheelchair navigation module and begin autonomous navigation. The wheelchair will avoid obstacles according to the environmental map and adjust the speed and direction to ensure safe navigation; Step 5: When the wheelchair reaches the target location, the user can get on and off easily. The system can automatically stop the vehicle according to the user's instructions or gestures, and help the user get on and off as needed.

[0030] Working principle: First, after the system is started, the visual sensor component module starts working to monitor the user's posture and position information. The camera captures the user's body posture and relative position information with the wheelchair. At the same time, the lidar and depth sensor detect the distance and relative position between the user and the wheelchair. These data are transmitted to the control and execution module for analysis and processing; the user can send a call signal through voice command or mobile phone application, which informs the system of the user's needs, such as the need to get on and off the car, move to a certain location, etc. After receiving the user's signal and sensor data, the motion control and wheelchair navigation module will formulate the best navigation route according to the user's needs and environmental conditions. Taking into account the obstacles in the environment, the navigation algorithm will determine a safe path to ensure that the wheelchair can reach the target location smoothly; once the navigation route is determined, the wheelchair body will start the motor according to the instructions of the motion control and wheelchair navigation module and start autonomous navigation. The wheelchair will avoid obstacles according to the environmental map and adjust the speed and direction to ensure safe navigation; when the wheelchair reaches the target location, the user can get on and off conveniently, and the system will automatically stop according to the user's instructions or gestures, and help the user get on and off as needed.

[0031] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent wheelchair automatic navigation system, characterized in that: It includes visual sensor module, control and execution module, motion control and wheelchair navigation module, user feedback and system optimization module, trajectory analysis and preference learning module and user interface module; The visual sensor component module includes but is not limited to a high-resolution camera, a laser radar, and a depth sensor, which are responsible for real-time monitoring of the environment around the wheelchair, capturing the user's posture information and position data, and transmitting the captured data information to the control execution module; The control and execution module includes a data processing and fusion unit and a path planning and navigation unit. The control and execution module receives sensor data information from the visual sensor component module, processes the information through the data processing and fusion unit, and transmits the processed information to the path planning and navigation unit. The path planning and navigation unit determines the best path for the wheelchair based on user needs, environmental information, and historical trajectory analysis and preferences, and transmits the path information to the motion control and wheelchair navigation module; The motion control and wheelchair navigation module includes a PID controller, a wheelchair body and an obstacle avoidance and dynamic adjustment unit. The wheelchair body is the physical part of the intelligent wheelchair, including an electric wheelchair chassis, a seat and a battery. These components work together and based on the path planning data of the path planning and navigation unit in the control and execution module, the wheelchair can control the wheelchair body to move along the path of the control and execution module according to the instructions of the PID controller. During the movement of the wheelchair body, the path route is dynamically adjusted by the obstacle avoidance and dynamic adjustment unit, and the data information is transmitted to the user feedback and system optimization module in real time; The user feedback and system optimization module collects real-time feedback from users through voice recognition, touch screen, buttons, etc., and uploads the feedback information to the motion control and wheelchair navigation module in real time. The motion control and wheelchair navigation module processes the received information and combines it with the real-time trajectory analysis and preference learning module to control the re-planning and design of the route until the wheelchair journey is completed; The historical trajectory analysis and preference learning module is based on the final data of the motion control and wheelchair navigation modules. The PDI controller transmits the data information of the path to the historical trajectory analysis and preference learning module. The system will regularly collect the user's historical travel trajectory data and use machine learning algorithms to identify the user's travel mode. The system can predict the user's future travel needs and recommend a route that is more in line with their habits. The user interface module is a way for users to interact with the system, usually including voice recognition, buttons, and mobile phone applications. Users can use the user interface module to control the movement status of the wheelchair in real time, view the walking trajectory route, and timely feedback and save data information such as the historical trajectory analysis and preference learning module, user feedback and system optimization module.

2. The intelligent wheelchair automatic navigation system according to claim 1, characterized in that: The visual sensor component module includes a high-resolution camera, a laser radar and a depth sensor; The camera is used to capture the user's visual information, including facial expressions, gestures, head posture, etc., so that the system can better understand the user's needs; The laser radar measures the distance and shape of the surrounding environment through the reflection of laser beams to build maps and detect obstacles; The distance sensor is used to obtain the distance and relative position between the user and the wheelchair to ensure safe boarding and alighting.

3. The intelligent wheelchair automatic navigation system according to claim 2, characterized in that: The control and execution module includes a data processing and fusion unit and a path planning and navigation unit. The data processing and fusion unit data processing analysis is as follows: The data processing and fusion unit data includes user position coordinate data processing and user sitting angle data processing: the user position coordinate data, that is, the environment image of the user and the wheelchair captured by the depth camera or laser radar, the user position coordinate is in the two-dimensional control key, and the user coordinate is recorded as , represents the coordinates of the user on the ground; the posture angle data can be obtained through sensor data, recorded as , the coordinate system of the wheelchair is The relative position of the user and the wheelchair can be calculated by rotation and translation. The rotation matrix Used to indicate changes in attitude angle: if is the coordinate of the user relative to the wheelchair (in the wheelchair coordinate system), then .

4. The intelligent wheelchair automatic navigation system according to claim 3, characterized in that: The data processing and fusion unit also includes a Kalman filter algorithm. The data processing and fusion unit processes the user's position coordinate data based on the dynamic environment and the changing posture and the user's sitting angle data, and uses algorithms such as Kalman filtering to continuously update and optimize the user's position and posture. The Kalman filter formula is as follows: in is the estimated state at the current moment, represents the observed sensor data, It represents the observation matrix; It represents the Kalman gain.

5. The intelligent wheelchair automatic navigation system according to claim 2, characterized in that: The path planning and navigation unit calculates the best path according to the data information sent by the control and execution module and the historical trajectory information of the historical trajectory analysis and preference learning module and the user's current position and target position. During the planning process, the system will continuously optimize the path, avoid obstacles and congested areas in real time; the path planning and navigation unit includes environment modeling and obstacle detection, path planning algorithm and path optimization. The environment modeling and obstacle detection are performed by laser radar, ultrasonic sensor or camera. The system can detect surrounding obstacles and record the obstacle set as: Each obstacle Each has its place The path planning algorithm is based on obstacle data and data information from the data processing and fusion unit. Algorithm, The algorithm uses heuristic search to find the starting point To the destination The shortest path of The cost function of the algorithm is: in represents the actual cost from the starting point to the current node n, is the heuristic estimated cost from the current node n to the target point, usually using the Euclidean distance: The path optimization, when the obstacle changes or the user's posture changes, the path needs to be dynamically adjusted, and the dynamic programming or online optimization algorithm is used to adjust the path as follows: in are the adjacent nodes on the path, is the Euclidean distance between two points.

6. The intelligent wheelchair automatic navigation system according to claim 1, characterized in that: The motion control and wheelchair navigation module includes a PID controller, a wheelchair body and an obstacle avoidance and dynamic adjustment unit. The PID controller is based on the data controlled by the control and execution module. The current position of the wheelchair is , the target location is The error is: Output control signal of PID controller for: in is the proportionality factor, used to adjust the error, is the integration factor, used to eliminate long-term errors , is the differential coefficient, used to predict the error change.

7. The intelligent wheelchair automatic navigation system according to claim 6, characterized in that: The obstacle avoidance and dynamic adjustment, during the wheelchair driving, obstacles may suddenly appear. In order to avoid collision, the wheelchair needs to monitor the surrounding environment in real time and adjust the path. The obstacle avoidance and dynamic adjustment include a dynamic window method, which includes a wheelchair body state space, a wheelchair body speed window, obstacle detection and evaluation, an evaluation function, and selection of the optimal speed. The dynamic window method calculates the current available speed window of the wheelchair body in real time, and then selects a suitable speed combination to avoid obstacles, and plans the optimal motion trajectory on this basis. The obstacle avoidance and dynamic adjustment are specifically analyzed as follows: The wheelchair main body state space, the current speed of the wheelchair main body is ,in is the linear velocity, is the angular velocity; The wheelchair body speed window is based on the data of the wheelchair body state space according to the maximum linear speed of the wheelchair body. and maximum angular velocity , the dynamic window defines the range of speeds that can be selected, and the boundaries of the window are usually: Line speed: Angular velocity: The obstacle detection and assessment uses sensors (such as lidar, ultrasonic sensors, etc.) to detect the distance of surrounding obstacles in real time, and calculates the collision risk of each speed combination based on the position of the obstacles; The evaluation function is for each candidate speed combination , calculate the following function To measure its quality: in Estimate whether the current speed combination helps to move toward the target point, which is usually a function related to the distance or direction of the target point; Evaluate whether the speed combination avoids the obstacle, usually based on the shortest distance to the obstacle; Consider the dynamic constraints of the wheelchair body to ensure that the speed combination does not exceed the dynamic limit of the wheelchair body. is the weight coefficient used to balance various evaluation factors; The optimal speed is selected for all candidate speed combinations. Select the evaluation value The maximum speed is taken as the final selected speed and used as the control input of the wheelchair.

8. The intelligent wheelchair automatic navigation system according to claim 7, characterized in that: The historical trajectory analysis and preference learning module is based on the final walking data of the motion control and wheelchair navigation modules. The PDI controller transmits the data information of the path to the historical trajectory analysis and preference learning module. The system will regularly collect the user's historical travel trajectory data and use machine learning algorithms to identify the user's travel pattern. The system can predict the user's future travel needs and recommend routes that are more in line with their habits.

9. The intelligent wheelchair automatic navigation system according to claim 8, characterized in that: The user interface module, the user interface is the way for users to interact with the system, usually including voice recognition, buttons, mobile phone applications, etc. Users can interact with the wheelchair in the following ways: The buttons and touch screen: The buttons or touch screen on the wheelchair allow users to send call signals, start navigation and provide feedback; The voice command: The user can tell the wheelchair their needs through voice commands, such as "forward", "stop", etc. The mobile phone application: The wheelchair can be connected to the user's smartphone application, and the user can monitor the wheelchair position and status in real time through the application, and send instructions.

10. The method based on an intelligent wheelchair automatic navigation system according to claim 9, characterized in that: Using the intelligent wheelchair automatic navigation system described in any one of claims 1 to 9, the steps are as follows: Step 1: First, after the system is started, the visual sensor component module starts working to monitor the user's posture and position information. The camera captures the user's body posture and relative position information with the wheelchair. At the same time, the laser radar and depth sensor detect the distance and relative position between the user and the wheelchair. These data are transmitted to the control and execution module for analysis and processing; Step 2: The user can send a summon signal through voice commands or mobile phone applications, which informs the system of the user's needs, such as needing to get on or off the vehicle, or move to a certain location. Step 3: After receiving the user's signals and sensor data, the motion control and wheelchair navigation module will develop the best navigation route based on the user's needs and environmental conditions. Taking into account obstacles in the environment, the navigation algorithm will determine a safe path to ensure that the wheelchair can reach the target location smoothly; Step 4: Once the navigation route is determined, the wheelchair body will start the motor according to the instructions of the motion control and wheelchair navigation module and begin autonomous navigation. The wheelchair will avoid obstacles according to the environmental map and adjust the speed and direction to ensure safe navigation; Step 5: When the wheelchair reaches the target location, the user can get on and off easily. The system can automatically stop the vehicle according to the user's instructions or gestures, and help the user get on and off as needed.

Images