TW201913599A - Automatic coach system and method in vehicle driving - Google Patents

Abstract

The present invention relates to an automatic coach system and an automatic coach method in vehicle driving. The system includes a central control unit, a voice input and output unit, a sensing navigation unit, a network interface unit electrically connected to the sensing navigation unit, a braking unit, an accelerator unit, a steering unit, a cloud supercomputer and a display unit. The method includes steps of positioning a vehicle by use of a global satellite navigation system to confirm latitude and longitude coordinates of the vehicle; selecting driving training items via the voice input and output unit; sensing movements of the vehicle to an initial point by use of the sensing navigation unit; and transmitting coach invoices in response to the selected driving training items by the cloud supercomputer and display the same via the invoice input and output unit. Accordingly, users may operate and carry out the driving program based on the coach voices. Besides, in case of any dangerous driving situations happened in the course of driving training, the central control unit is served to commend the braking, accelerator and steering units to do proper responses to the actual dangerous driving conditions, so as further to avoid dangerous driving events.

Description

Automated trainer system and method for vehicle driving

The present invention relates to an automated instructor system for driving a vehicle, and more particularly to an instructor who uses artificial intelligence and an autonomous vehicle as a driving class to teach the participants to learn various road test items in the vehicle examination.

At present, in addition to the written test, all kinds of procedures for obtaining a driver's license must pass the road test. The road test items are: reverse storage, parallel roadside parking, curve advance and retreat, Quxiang turn head, railway level crossing, crossroad crossing, shifting Stable test, up and down ramp, narrow bridge, zebra pattern pedestrian crossing line, full road driving; if the road test project is pressed into the pipeline, the deduction standard is usually 16 points, and the pass score is 70 points, so only twice Pressed into the pipeline, the road test failed, and the driver's license could not be obtained. It must be re-examined. For each of the above road test projects, each driving course itself has a set of standard driving skills that can be easily passed. Therefore, the test taker usually participates in the driving learning course of the driving class to teach driving skills through the driving class. Successfully passed various road test projects.

However, at present, the driving training class teaches and masters the driving skills of the road test. The driving skills are taught by the real coach. That is, each time and each student must have a coach. In this case, the number of students in the same stage will be Restricted the number of coaches who can attend the teaching skills, which is not conducive to the business expansion of the driving class.

Nowadays, the present inventors have further provided an improved technique for the deficiencies of the existing situation in view of the above-mentioned situation, so that the number of trainees at the same time is no longer limited, and the number of coaches who can teach skills can be attended.

The main object of the present invention is to provide an automated coaching system and method for driving a vehicle, which mainly teaches the trainer by combining artificial intelligence and automatic driving technology, and teaching the coach as a skill of various road test projects of the driving class. Learn the driving skills of various road test projects.

The object and effect of the present invention are achieved by the following techniques:

An automated coaching system for vehicle driving includes a central processing unit, a voice input/output unit, a sensing/navigation unit, and a network interface unit; wherein:

The central processing unit is electrically connected to the voice input/output unit, the sensing/navigation unit, the network interface unit, and a cloud supercomputer, and the central processing unit is further coupled to a vehicle The unit, the throttle unit, the steering unit, and the display unit are electrically connected; the central processing unit may control an operation timing and a mode of the brake unit, the throttle unit, the steering unit, and the display unit; a voice input by the voice input/output unit and a sensing result of the sensing/navigation unit, and uploading the input voice and the sensing result to the cloud super computer via the network interface unit, and The network interface unit returns the execution result of the driving instructor system program of the cloud super computer to the central processing unit, and the central processing unit performs voice and image output, and control of the brake, the throttle, and the steering wheel;

The cloud supercomputer has built-in teaching skills for various road test projects, and has a software program that performs navigation and dangerous driving judgment functions, and can select a corresponding road test item skill to teach voice according to the input voice. And outputting to the voice input/output unit for playing;

The voice input/output unit can receive a road test item of the student voice input, and can send a voice output from the cloud super computer to the road test item of the central processing unit to the voice input/output unit to play;

The sensing/navigation unit senses the surrounding environment and location of the vehicle through a sensor mounted on the vehicle body, and transmits the sensing result to the cloud super computer through the network processing unit through the central processing unit. Determining, by the cloud supercomputer, the vehicle position and the vehicle as normal driving or dangerous driving according to the sensing results, and then the central processing unit controls the braking unit, the throttle unit, and the center according to the judgment result of the cloud supercomputer. The steering wheel unit performs a corresponding dynamic sample, and the cloud supercomputer can plan a route data according to the road test item input by the student voice, and display it on the display unit to guide the trainer to drive the vehicle to the road test. Starting position

The network interface unit is configured to connect the central processing unit to the cloud supercomputer to send voice input from the voice input/output unit to the central processing unit to the The cloud supercomputer performs speech recognition and language processing, and can transmit the result of the cloud supercomputer judgment and processing to the central processing unit via the network interface unit, so that the central processing unit controls the result according to the result. a voice input/output unit, the sensing/navigation unit, the brake unit, the throttle unit, and the steering unit;

According to this, the trainees of the driving training can select the road test project through the voice to instantly play the voice corresponding to the driving skills of the road test project, and continuously check whether there is a dangerous driving situation during the driving of the trainer, when the cloud When the supercomputer determines that it is dangerous driving, the central processing unit controls the braking unit, the throttle unit, and the steering unit to change the vehicle from the human driving mode to the automatic driving mode.

The automatic coaching system for driving a vehicle as described above, wherein the road test project includes reverse storage, parallel roadside parking, curve advance and retreat, curved lane turning, railway level crossing, crossroad crossing, (hand row) shifting Stable test, up and down ramp, narrow bridge, zebra pattern pedestrian crossing line, full road driving.

An automated coaching method for vehicle driving, comprising the following steps

Step 1: Vehicle positioning, obtaining the initial latitude and longitude coordinates of the vehicle through the global satellite navigation system;

Step 2: Select the road test project, and the driver selects the road test project for driving training;

Step 3: Navigate to the starting point, and after completing the selection of the road test project, guide the vehicle to the starting point of the road test project by navigation;

Step 4: Play the training driving skill voice. When the vehicle reaches the starting point and starts driving training for the road test project, the training driving skill voice of the road test item is played first;

Step 5: judging whether it is dangerous driving, when the driver starts driving training of the road test item, determining whether the current driving state of the vehicle is dangerous driving according to the driving data of the vehicle; if yes, proceeding to step 6; if not, proceeding to step 7;

Step 6: Automatic driving to take over, after judging that the vehicle is dangerously driven, the vehicle starts the automatic driving mode, and the vehicle independently controls the brake, the throttle and the steering wheel, and automatically drives back to the starting point of the road test item;

Step 7: Improve the situation, analyze and evaluate the actual driving situation of the driver's road test project, and inform the place to be improved.

The automatic coaching method for vehicle driving as described above, wherein the road test item includes reverse storage, parallel roadside parking, curve advance and retreat, curved lane turning, railway level crossing, crossroad crossing, (hand row) shifting Stable test, up and down ramp, narrow bridge, zebra pattern pedestrian crossing line, full road driving.

The advantages of the invention are:

The automated coaching system and method for driving a vehicle of the present invention can connect artificial driving to teach artificial intelligence how to drive through artificial intelligence. When the student drives incorrectly, artificial intelligence can assist in correcting and timely teach driving skills.

The automated instructor system and method for driving a vehicle of the present invention is usually driven by a driver, but in the event of a critical situation of dangerous driving, the vehicle is automatically driven by the vehicle.

The automated instructor system and method for vehicle driving of the present invention uses voice to communicate with the driver and teaches the student how to drive.

For a more complete and clear disclosure of the technical content, the purpose of the invention and the effects thereof achieved by the present invention, it is explained in detail below, and please refer to the drawings and drawings:

Please refer to the first picture.

The automated coaching system for driving a vehicle of the present invention comprises: a central processing unit (1), a voice input/output unit (2), a sensing/navigation unit (3), and a network interface unit (4); :

The central processing unit (1) is electrically connected to the voice input/output unit (2), the sensing/navigation unit (3), the network interface unit (4), and the central processing unit (1) is also coupled to the braking unit of the vehicle ( 5) The throttle unit (6), the steering unit (7), and the display unit (9) are electrically connected; the central processing unit (1) receives the voice input from the voice input/output unit (2) and the sensing/navigation unit (3) After the sensing result, the received voice and the sensing result are uploaded to a cloud super computer (8) via the network interface unit (4), and the cloud super computer (8) performs calculation according to the voice and the sensing result. The processing is judged, and the result of the processing by the cloud supercomputer (8) is returned to the central processing unit (1) by the network interface unit (4), and the central processing unit (1) controls the voice input/output unit according to the processing result ( 2), the timing and mode of operation of the brake unit (5), the throttle unit (6), and the steering wheel unit (7), and can guide the vehicle to a preset position;

The voice input/output unit (2) mainly receives the road test item of the student's voice input, and the voice input test item is uploaded to the cloud super computer (8) through the central processing unit (1) through the network interface unit (4). The cloud supercomputer (8) then transmits the corresponding teaching method of the road test item to the central processing unit (1) according to the voice input project, and is output by the central processing unit (1). To the voice input/output unit (2) to play the skills of the road test project to teach the voice, for the driver to teach the voice according to the skill of the road test project for actual driving operation training;

The sensing/navigation unit (3) mainly comprises at least one sensor (such as a distance sensor, a GPS module, a radar, an optical radar) mounted on the vehicle, and sensing the surrounding environment and position of the vehicle through the sensor, such as the present Whether the vehicle maintains a safe distance between the vehicle and the vehicle, whether the vehicle deviates from the road, the current position of the vehicle...), and transmits the sensing result to the cloud supercomputer through the central processing unit (1) through the network interface unit (4) (8) ), the cloud supercomputer (8) judges the vehicle position and the vehicle as normal driving or dangerous driving according to the sensing results, and then the central processing unit (1) controls the braking unit according to the judgment result of the cloud supercomputer (8) (5) ), the throttle unit (6), the steering wheel unit (7) performs a corresponding dynamic sample, and the cloud supercomputer (8) plans a route information according to the student's voice input road test project, and the route information is via the network interface unit ( 4) Displayed with the central processing unit (1) to the display unit (9) of the vehicle to guide the trainer to drive the vehicle to the road Starting position;

The network interface unit (4) is a network module, and the network connection mode can be 4G, 5G or WiFi. The network interface unit (4) is between the central processing unit (1) and the cloud supercomputer (8). The connection interface can send the voice input from the voice input/output unit (2) into the central processing unit (1) to the cloud supercomputer (8) for voice recognition, language processing, and judging and processing the cloud supercomputer (8). The result is transmitted to the central processing unit (1), and the central processing unit (1) controls the voice input/output unit (2), the sensing/navigation unit (3), the braking unit (5), and the throttle unit according to the result. ), the steering wheel unit (7) operates.

That is, the driving training student selects the learning road test item by voice through the voice input/output unit (2). The voice message is received by the central processing unit (1) and then uploaded to the cloud super computer via the network interface unit (4) ( 8) The cloud supercomputer (8) then returns the voice corresponding to the driving skill of the road test project to the central processing unit (1) according to the selected road test project via the network interface unit (4), and then the central processing unit (1) It is sent to the voice input/output unit (2) for playing. At this time, the driver can perform on-the-spot driving training according to the driving skill voice; while the trainer is driving, the sensing/navigation unit (3) continuously senses the vehicle. Location and surrounding environment, and the sensing results are uploaded to the cloud supercomputer (8) via the central processing unit (1) and the network interface unit (4), and the cloud supercomputer (8) also performs driving safety according to the sensing results. Judgment, check whether there is a dangerous driving situation, when it is determined that it is dangerous driving, the cloud super computer (8) will make the central processing Element (1) to take over the braking control unit (5), an accelerator unit (6), a steering wheel unit (7), the driving of the vehicle by human driving mode to automatic mode.

In particular, the system of the present invention can be used to implement a method of automated coaching for vehicle driving, the method comprising the steps of:

Step 1: Vehicle positioning, obtaining the initial latitude and longitude coordinates of the vehicle through the global satellite navigation system;

Step 2: Select a road test item, and the driver selects a road test item for driving training. In this embodiment, the driver selects a road test item by voice through a voice input/output unit (2);

Step 3: Navigate to the starting point. After completing the selection of the road test project, navigate the vehicle to the starting point of the road test project by navigation;

Step 4: Play the training driving skill voice. When the vehicle reaches the starting point and starts driving training for the road test project, the training driving skill voice of the road test item is played first;

Step 5: judging whether it is dangerous driving, when the driver starts driving training of the road test item, determining whether the current driving state of the vehicle is dangerous driving according to the driving data of the vehicle; if yes, proceeding to step 6; if not, proceeding to step 7;

Step 6: Automatic driving to take over, after judging that the vehicle is dangerously driven, the vehicle starts the automatic driving mode, and the vehicle independently controls the brake, the throttle and the steering wheel, and automatically drives back to the starting point of the road test item;

Step 7: Improve the situation, analyze and evaluate the actual driving situation of the driver's road test project, and inform the place to be improved.

At present, vehicle road test projects include: reversing storage, roadside parking, curve advance and retreat, Quxiang turning head, railway level crossing, crossroad crossing, (hand row) shift stability test, up and down ramp, narrow bridge, zebra pattern pedestrian crossing Road line, full road driving; the following is further explained according to the actual operating conditions.

Please refer to the second figure, which is applied to the reverse storage, parallel roadside parking, curve advance and retreat, curved lane turning, railway level crossing, crossroad crossing, (hand row) shift stability test, up and down ramp, narrow bridge Flowchart for various road test items such as zebra crossing pedestrian crossing or full road driving.

First of all, the driving trainer selects the road test project first, in order to explain the following road test items that select “reverse storage”, then locate the location of the vehicle and guide the vehicle to the starting point of the road test project. Ask if the driving trainees start. If they do not respond or respond to "No", the system will delay asking for a period of time to ask if the driving trainees start, until the driving trainees respond to "Yes", then the voice input/output unit (2) will The "Reversing" driving skill voice is played. The driving trainees can perform driving according to the "Reversing Warehousing" driving skill voice, and promptly remind and correct the driving operation of the driving trainer "reverse storage", and further detect whether there is danger. The driving situation occurs. If it is judged as "Yes", the vehicle will start the automatic driving mode and automatically drive the vehicle to the starting point of "Reversing storage". If it is judged as "None", it is further judged whether or not "Reversing storage" is completed. In the road test project, if the judgment is not completed, the system will return to the voice input/output unit (2). Reversing storage state "driving skills of speech, if it is judged complete, the system further integration of areas for improvement, and inform the driving training students.

The above is only a part of the embodiments of the present invention, and is not intended to limit the present invention. It is intended to be included in the scope of the present invention.

In summary, the embodiments of the present invention can achieve the expected use efficiency, and the specific technical means disclosed therein have not been seen in similar products, nor have they been disclosed before the application, and have completely complied with the patent law. The regulations and requirements, the application for invention patents in accordance with the law, and the application for review, and the grant of patents, are truly sensible.

(1)‧‧‧Central Processing Unit

(2) ‧‧‧Voice input/output unit

(3)‧‧‧Sensing/navigation unit

(4) ‧‧‧Network interface unit

(5) ‧‧‧ brake unit

(6)‧‧‧ throttle unit

(7)‧‧‧Steering wheel unit

(8) ‧‧‧Cloud Supercomputer

(9)‧‧‧Display unit

First figure: Schematic diagram of the control block architecture of the present invention.

Second Figure: Flow chart of the steps of the present invention.

Claims (4)

An automated coaching system for driving a vehicle, comprising a central processing unit, a voice input/output unit, a sensing/navigation unit; wherein: the central processing unit and the voice input/output unit, the sensing/navigation The central processing unit is electrically connected to the braking unit, the throttle unit, the steering unit and the display unit of the vehicle; the central processing unit can control the braking unit and the a throttle unit, the steering wheel unit, an operation timing and mode of the display unit; and capable of receiving a voice input from the voice input/output unit and a sensing result of the sensing/navigation unit, and inputting The voice and the sensing result are uploaded to the cloud supercomputer through the network interface unit, and the execution result of the driving supervisor system program of the cloud super computer can be returned to the central processing unit through the network interface unit. , the central processing unit performs voice and video output, and control of the brake, throttle, and steering wheel; The cloud supercomputer has built-in teaching skills for various road test projects, and has a software program that performs navigation and dangerous driving judgment functions, and can select a corresponding road test item skill to teach voice according to the input voice. And outputting to the voice input/output unit for playing; the voice input/output unit can receive the road test item of the student voice input, and can transmit the road test item from the cloud super computer to the central processing unit a technique for teaching voice output to the voice input/output unit to play; the sensing/navigation unit sensing a surrounding environment and location of the vehicle through a sensor mounted on the vehicle body, and transmitting the sensing result to the central processing unit The cloud supercomputer transmits to the cloud super computer, and the cloud super computer determines the location of the vehicle and the vehicle for normal driving or dangerous driving according to the sensing results, and then the central processing unit is configured according to the cloud super computer. Determining a result, controlling the brake unit, the throttle unit, and the steering wheel The unit performs a corresponding dynamic sample, and the cloud super computer can plan a route information according to the road test item input by the student voice, and display the information on the display unit to guide the trainer to drive the vehicle to the start position of the road test. The network interface unit is configured to connect the central processing unit to the cloud supercomputer to send voice input from the voice input/output unit to the central processing unit to the network through the network interface unit The cloud supercomputer performs speech recognition and language processing, and can transmit the result of the cloud supercomputer judgment and processing to the central processing unit via the network interface unit, so that the central processing unit controls the location according to the result. The voice input/output unit, the sensing/navigation unit, the brake unit, the throttle unit, the steering wheel unit, and the display unit operate; accordingly, the driving training student can learn by voice selection Road test project to instantly play the voice of the driving skills of the road test project, and in the process of driving students Continuing to check whether there is a dangerous driving situation, when the cloud supercomputer determines that it is dangerous driving, the central processing unit controls the braking unit, the throttle unit, and the steering wheel unit to make the vehicle drive by humans. The mode is changed to the automatic driving mode. The automatic instructor system for driving a vehicle according to claim 1, wherein the road test item comprises a reverse storage, a parallel roadside parking, a curve advancing and retreating, a curved lane turning, a railway crossing, a crossroad crossing, ( Hand-cranked car) shift stability test, up and down ramp, narrow bridge, zebra pattern pedestrian crossing line, full road driving. An automated coaching method for driving a vehicle, comprising the following steps: Step 1: Vehicle positioning, obtaining an initial latitude and longitude coordinate of the vehicle through a global satellite navigation system; Step 2: selecting a road test item, and selecting a road test item for driving training by the driver Step 3: Navigate to the starting point, after selecting the road test project, guide the vehicle to the starting point of the road test project by navigation; Step 4: Play the training driving skill voice, when the vehicle reaches the starting point, and start Before the driving training of the road test project, the training driving skill voice of the road test item is played first; Step 5: determining whether the driving is dangerous, and when the driver starts driving training of the road test item, the current driving state of the vehicle is determined according to the driving data of the vehicle. Whether it is dangerous driving; if yes, go to step 6; if not, go to step 7; Step 6: Automatic driving to take over, after judging that the vehicle is dangerous driving, the vehicle starts the automatic driving mode, and the vehicle controls the brakes and throttles autonomously. Steering wheel and self Driving back to the starting point of the road test project; Step 7: improve matters, analyze and evaluate actual driving conditions for motorists the road test project and inform areas for improvement. The automatic coaching method for vehicle driving according to claim 3, wherein the road test item includes reverse storage, parallel roadside parking, curve advance and retreat, curved lane turning, railway level crossing, intersection, ( Hand-cranked car) shift stability test, up and down ramp, narrow bridge, zebra pattern pedestrian crossing line, full road driving.