Disclosure of Invention
Aiming at the problems in the prior art, the invention provides the blind guiding robot dog based on artificial intelligence, which does not need manpower and time consumption training and post care. The blind person can conveniently go out in daily life, and the blind person navigation system has the functions of obstacle avoidance, navigation, traffic light identification and the like. Meanwhile, the problem that the blind person cannot go up with the guide dog in some areas such as subways and buses is effectively avoided.
In order to achieve the purpose, the invention adopts the technical scheme that: a blind guiding machine dog based on artificial intelligence comprises a pulling rope and a blind guiding machine dog body connected with one side of the pulling rope, wherein a main control module is arranged in the blind guiding machine dog body; the main control module is connected with a communication module, a driving module and a voice interaction module; the communication module comprises a Bluetooth headset; the main control module also comprises a perception module, a collision avoidance following module, a vision module and a navigation module;
the sensing module comprises a light band sensor which can detect the existence of the blind road and guide a user to walk on the blind road by sensing the ground flatness through a photoelectric sensor;
the collision avoidance following module comprises a laser radar, and data is processed through a single chip microcomputer control system, so that the target direction of a user is ensured to be accurate;
the navigation module consists of a gyroscope, a GPS positioning module, a map module and a path planning module, and has the functions of GPS satellite positioning, electronic map browsing and inquiring and intelligent path planning, wherein the voice prompt of the navigation system is controlled by a special voice synthesis subsystem;
the driving module comprises a motor and a roller, and the roller is controlled to move under the action of the control unit through the motor;
the vision module comprises a camera and an image processing module; the camera collects data such as traffic lights and zebra crossings, the transmitted image data is subjected to depth processing through the image processing module, so that the traffic lights and the zebra crossings are judged, the judgment result is transmitted to the main control module, and a user is informed through the voice interaction module.
Furthermore, the voice interaction module has the functions of voice operation and voice prompt, when the voice interaction module is used, when a user sends a voice instruction, a voice recognition engine in the voice interaction module can obtain a result according to the current command word and transmit the result to the main control module, and therefore the next step of instruction is carried out.
Furthermore, the navigation module adopts laser SLAM to position and navigate
Further, when the camera detects a traffic light, the image is transmitted back to the image processing module, the traffic light is framed and selected according to the Yolov3 target, and meanwhile, the traffic light condition is monitored in real time; when the red light is detected, the information is transmitted to the main control module, the main control module controls the driving module, the driving motor stops rotating, and the machine dog stops; meanwhile, the voice module sends out a prompt and transmits the prompt to the Bluetooth headset through the communication module to prompt a user to be red light and stop acting; when the green light is detected, the robot dog continues to act, and meanwhile, the voice module also reminds the robot dog of the green light and sends out different calls; when the user crosses the road, the main control module controls the obstacle avoidance radar to continuously work, and when the fact that the distance of the running vehicle is too close is detected, the voice module directly sends out a warning sound to remind the running vehicle, and meanwhile the user is guaranteed to cross the road safely.
Furthermore, the sensing module can detect the existence of the blind road through the sensing ground flatness of the photoelectric sensor and guide a user to walk on the blind road; when a user walks on the blind road, the photoelectric sensors at two sides of the robot dog perform detection once every other time period, and when the robot dog detects that the robot dog is not in the blind road or is about to deviate from the blind road, the main control module controls the driving module and the driving motor performs corresponding adjustment; when the right side is detected to leave the blind road soon, the current of the motor on the right side is increased, and the current of the motor on the left side is reduced, so that the blind guiding machine dog returns to the center of the blind road again.
Furthermore, the obstacle avoidance following module can process data through a control system to ensure the target direction of a user; the laser radar monitors in real time, and can control the rotating speed of the motor according to the walking speed of a user, so that the walking speed of the robot dog is controlled, and the distance between the user and the blind guide robot dog is controlled within a certain safety range; when the obstacle avoidance function is performed, the laser radar transmits a detection signal every few seconds, then the received signal reflected from the target is compared with the transmitted signal, and the distance between the user and the target is determined by measuring the time of the laser signal in back-and-forth propagation.
Furthermore, a safe distance is set in advance by the obstacle avoidance following module, and when the distance is smaller than the safe distance, the main control module directly sends out a large alarm sound through the voice interaction module to remind a user and also remind passersby; when the vehicle passes through the road, the vehicle can be reminded to present the user in advance.
The invention has the beneficial effects that: the designed blind guiding robot dog does not need long-time training, so that the consumption of manpower and time is reduced; the blind guide robot dog does not need the care of a user, and is more convenient; the guide dog robot can assist a user to get on a subway, a public transport and the like, and the problem that the blind cannot be carried on the guide dog in the public transport and the subway in partial areas is solved; the blind guiding robot dog can assist a user in going out by distinguishing traffic lights, zebra crossings and blind roads.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood, however, that the description herein of specific embodiments is only intended to illustrate the invention and not to limit the scope of the invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, and the terms used herein in the specification of the present invention are for the purpose of describing particular embodiments only and are not intended to limit the present invention.
As shown in fig. 1, the artificial intelligence-based blind guiding robot dog comprises a pulling rope and a blind guiding robot dog body connected to one side of the pulling rope, wherein a main control module is arranged in the blind guiding robot dog body; the main control module is connected with a communication module, a driving module and a voice interaction module; the communication module comprises a Bluetooth headset; the main control module also comprises a perception module, a collision avoidance following module, a vision module and a navigation module;
the sensing module comprises a light band sensor which can detect the existence of the blind road and guide a user to walk on the blind road by sensing the ground flatness through a photoelectric sensor;
the collision avoidance following module comprises a laser radar, and data is processed through a single chip microcomputer control system, so that the target direction of a user is ensured to be accurate;
the navigation module consists of a gyroscope, a GPS positioning module, a map module and a path planning module, and has the functions of GPS satellite positioning, electronic map browsing and inquiring and intelligent path planning, wherein the voice prompt of the navigation system is controlled by a special voice synthesis subsystem;
the driving module comprises a motor and a roller, and the roller is controlled to move under the action of the control unit through the motor;
the vision module comprises a camera and an image processing module; the camera collects data such as traffic lights and zebra crossings, the transmitted image data is subjected to depth processing through the image processing module, so that the traffic lights and the zebra crossings are judged, the judgment result is transmitted to the main control module, and a user is informed through the voice interaction module.
Preferably, the voice interaction module has the functions of voice operation and voice prompt, and when the voice interaction module is used, when a user sends a voice instruction, the voice recognition engine in the voice interaction module can obtain a result according to the current command word and transmit the result to the main control module, so that the next step of instruction is carried out.
Preferably, in this embodiment, the navigation module uses a laser SLAM to perform positioning navigation.
Referring to fig. 2, the working principle of the present invention is as follows: when a user sends an instruction to a specified place, a voice recognition engine of the voice interaction module can recognize a result according to a current command word, and the management window acquires a control command of the recognition result and then sends the control command to the navigation module. The modules communicate with each other through interfaces. The navigation module adopts laser SLAM positioning navigation, so that the navigation, the positioning and the path planning are more accurate and accurate.
The vision module comprises a camera and an image processing module. The camera collects data such as traffic lights, zebra crossings and the like, the transmitted image data is subjected to depth processing through the image processing module, so that the traffic lights and the zebra crossings are judged, the judgment result is transmitted to the main control module, and a user is informed through the voice interaction module.
When the camera of the machine dog detects the traffic light, the image is transmitted back to the image processing module, the traffic light is framed and selected according to the Yolov3 target, and meanwhile, the traffic light condition is monitored in real time. When the red light is detected, the information is transmitted to the main control module, the main control module controls the driving module, the driving motor stops rotating, and the machine dog stops. Meanwhile, the voice module sends out a prompt and transmits the prompt to the Bluetooth headset through the communication module to prompt a user to be red light and stop acting; when the green light is detected, the robot dog continues to act, and meanwhile, the voice module also reminds the robot dog of the green light and sends out different calls. When the user crosses the road, the main control module controls the obstacle avoidance radar to continuously work, and when the fact that the distance of the running vehicle is too close is detected, the voice module directly sends out a warning sound to remind the running vehicle, and meanwhile the user is guaranteed to cross the road safely.
Meanwhile, when the user waits for the bus at the station, the voice gives a command to wait for the bus. The voice interaction module recognizes the number and transmits the number to the main control module. The robot dog stops to camera real-time supervision is through public transit, and the line number on the bus is framed to the target, and when monitoring this line number, the instruction number that transmits among image processing module and the main control module contrasts, if contrast is correct, sends the instruction to the main control module in, and the pronunciation interactive module is passed through communication module simultaneously and is transmitted in the bluetooth headset, tells that the user bus has arrived at a station. A
The sensing module consists of a photoelectric sensor, and can detect the existence of the blind road and guide a user to walk on the blind road by sensing the ground flatness through the photoelectric sensor. When a user walks on the blind road, the photoelectric sensors on two sides of the machine dog detect every ten seconds, and when the machine dog detects that the machine dog is not on the blind road or is about to deviate from the blind road, the main control module controls the driving module and the driving motor performs corresponding adjustment. For example, when the right side is detected to leave the blind road soon, the current of the motor on the right side is increased, and the current of the motor on the left side is reduced, so that the blind guiding robot dog returns to the center of the blind road again.
The obstacle avoidance following module is composed of a laser radar. The function of avoiding obstacles and the function of automatically following the user are realized. The single chip microcomputer can process data through the control system, and the target direction of a user is guaranteed. Laser radar real-time supervision can come the rotational speed of control motor according to the speed of user's walking speed to the walking speed of control machine dog, thus the distance of control user and guide blind machine dog is in certain safety range. When the obstacle avoidance function is performed, the laser radar transmits detection signals every few seconds, then the received signals reflected from the target are compared with the transmitted signals, and the distance between a user and the target is determined by measuring the time of the laser signals in back-and-forth propagation, so that the obstacle avoidance function is achieved. A safe distance is set in advance, when the safe distance is smaller than the safe distance, the main control module directly sends out a large alarm sound through the voice interaction module, and a user is reminded of passing pedestrians. Particularly, when the vehicle passes through the road, the vehicle can be reminded to give the user a gift.
The designed blind guiding robot dog does not need long-time training, so that the consumption of manpower and time is reduced; the blind guide robot dog does not need the care of a user, and is more convenient; the guide dog robot can assist a user to get on a subway, a public transport and the like, and the problem that the blind cannot be carried on the guide dog in the public transport and the subway in partial areas is solved; the blind guiding robot dog can assist a user in going out by distinguishing traffic lights, zebra crossings and blind roads.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents or improvements made within the spirit and principle of the present invention should be included in the scope of the present invention.