TWI747651B - Autonomous driving assistant system - Google Patents

Abstract

An autonomous driving assistant system is provided to solve the problem that the position unit of the conventional vehicle has position inaccurate to cause positioning errors. The system includes a position unit for generating a vehicle coordinate. A correction module for obtaining correction parameter between a vehicle and a positioning reference. A processor module for receiving a parking path composed of a plurality of locus coordinates. There is an actual value between the each of the plurality of locus coordinates and the positioning reference. The processor module obtains a locus coordinate closest to the vehicle coordinate and controls the vehicle moves to the vehicle coordinate when the vehicle coordinate is not the same as a destination coordinate. The processor module calculates a difference between the correction parameter and the actual value and controls the vehicle movement by an electronic control unit of the vehicle according to the difference.

Description

Autonomous driving assistance system

The present invention relates to a vehicle auxiliary system, in particular, an automatic driving auxiliary system capable of correcting the position coordinates of a vehicle that is performing an automatic parking function.

Please refer to Figure 1, which is a conventional fully automatic parking path determining device 9. The conventional fully automatic parking path determining device 9 has a parking space module 91, a first determining module 92, and a first determining module. A second judgment module 93, a third judgment module 94, a route module 95, and a control module 96. The control module 96 determines a driveable area and a target parking space of a target vehicle through the parking module 91 , And execute the first judgment module 92 to judge whether the target vehicle can drive to the target parking space based on a first preset trajectory. If the judgment result is no, the second judgment module 93 is further executed to confirm Whether the target vehicle can drive to the target parking space based on a second preset trajectory, if the confirmation result is no, the third judgment module 94 is further executed to evaluate whether the target vehicle can be based on a third preset trajectory Drive to the target parking space, and if the evaluation result is yes, execute the path module 95 to set the third preset trajectory as the parking path of the target vehicle. A fully automatic parking route determination device 9 similar to the prior art has been disclosed in the Chinese Patent Publication No. 111497829.

In the above-mentioned conventional fully automatic parking path determining device 9, the control module 96 executes parking path planning through the first determining module 92, the second determining module 93, and the third determining module 94 in sequence After that, a positioning unit 97 can be used to detect and update the position of the target vehicle after movement. However, due to the problem of the accuracy of the positioning unit 97, the position coordinates detected by the positioning unit 97 will have slight errors. These errors will occur as the target vehicle moves along the parking path. The continuous cumulative growth resulted in the target vehicle being unable to stop correctly in the target parking space.

In view of this, the conventional fully automatic parking path determination device does still have to be improved.

In order to solve the above-mentioned problems, the object of the present invention is to provide an automatic driving assistance system that can correct the position coordinates of a vehicle that is performing an automatic parking function.

The second purpose of the present invention is to provide an automatic driving assistance system that can provide a driver with a remote call to the vehicle so that the vehicle can be automatically driven to the driver's side.

The directionality described in the full text of the present invention or its similar terms, such as "front", "rear", "left", "right", "up (top)", "down (bottom)", "inner", "outer" , "Side", etc., mainly refer to the directions of the attached drawings. Each directionality or similar terms are only used to help explain and understand the embodiments of the present invention, and are not used to limit the present invention.

The elements and components described in the full text of the present invention use the quantifiers of "one" or "one", which is only for convenience and to provide the general meaning of the scope of the present invention; in the present invention, it should be interpreted as including one or at least one, and single The concept of also includes the plural, unless it clearly implies other meanings.

The "Cloud Server" mentioned in the full text of the present invention is created by using virtualization software to divide a physical (bare metal) server into several virtual servers for application execution and information processing and storage; Users can remotely access the functions of these virtual servers through an online interface.

The "Database" mentioned in the full text of the present invention refers to a group of related The electronic data collection and storage in the hard disk, memory or a combination of the above, and can use the syntax function provided by the database management system (DBSMS), such as adding, reading, searching, updating and deleting, etc., to the electronic Data is processed; the database management system can manage electronic data in different data structures, for example, associative, hierarchical, mesh, or object-oriented, etc. The present invention uses a relational database management system as Examples are described below, but they are not intended to limit the present invention.

The "Coupling" mentioned in the full text of the present invention means that two devices can transmit data to each other by any direct or indirect connection means. For example, if the first device is coupled to the second device, it should be interpreted as that the first device can be directly connected to the second device in the present invention. Twisted pair) connection; or the first device can be indirectly connected to the second device through other devices or some kind of connection means, such as wireless media (such as: WiFi, Bluetooth) or heterogeneous network (Heterogeneous Network ) Connection. Those who have general knowledge in the field can choose according to the normal connection means of the device to be connected.

The automatic driving assistance system of the present invention includes: a positioning unit for installing on a vehicle, the positioning unit for locating the position of the vehicle to generate a vehicle body coordinate; a calibration module for installing on the vehicle, the The calibration module is used to obtain a reference distance and a reference angle between the vehicle and a positioning reference object; and a processor module electrically connected to the positioning unit and the calibration module, and the processor module receives When a parking signal is activated, the positioning unit and the calibration module are controlled to activate and receive a parking path. The parking path is composed of several track coordinates including an end point coordinate. A relative distance and a relative angle. The processor module analyzes whether the vehicle body coordinates are the same as the end point coordinates to confirm whether the vehicle is located in a parking space. If the analysis result is no, then the several track coordinates Obtain a track coordinate closest to the vehicle body coordinate, and the track coordinate is located between the vehicle and the parking space, the processor module sends a movement signal to an on-board computer of the vehicle, so that the on-board computer controls the The vehicle moves until the fixed The vehicle body coordinates detected by the bit unit are the same as the track coordinates. The processor module calculates the reference distance and the reference angle detected by the calibration module to obtain a calibration parameter, and calculates the calibration parameter and the The difference between the relative distance corresponding to the trajectory coordinates and an actual value calculated by the relative angle. If the difference is not equal to zero, a motion compensation signal is sent to the on-board computer so that the on-board computer controls the vehicle according to the difference. Perform movement compensation.

Accordingly, the automatic driving assistance system of the present invention can subdivide the parking path into several trajectory coordinates through a small number of hardware devices and simple computer processing operations, and make the vehicle only need to move to the next trajectory during each movement. The coordinates, and when the vehicle body coordinates detected by the positioning unit are the same as the track coordinates, the processor module obtains the reference distance and the reference angle between the vehicle and the positioning reference object through the calibration module , To obtain the correction parameter, and calculate the difference between the relative distance and the actual value calculated by the relative angle corresponding to the correction parameter and the trajectory coordinates, and when the difference is not equal to zero, make the on-board computer based on the difference The value controls the vehicle to perform motion compensation. In this way, the automatic driving assistance system of the present invention has the advantages of simplifying hardware costs, reducing computer computing performance requirements, completing path planning in a guided manner through the several trajectory coordinates, and improving the coordinate accuracy of the positioning unit through the calibration module And other effects.

Wherein, the several trajectory coordinates have a point coordinate. Before the vehicle moves along the several trajectory coordinates, the processor module can calculate the distance between the vehicle body coordinate and each trajectory coordinate, and compare it with The closest track coordinate of the vehicle body coordinate is used as the starting point coordinate. The processor module calculates a direction angle and a distance value of the vehicle body coordinate relative to the starting point coordinate, and sends the direction angle and the distance value including the direction angle and the distance value. A movement signal is sent to the vehicle-mounted computer, so that the vehicle-mounted computer can control the vehicle to move to the starting point coordinate according to the direction angle and the distance value. In this way, it has the function of correcting the initial position of the vehicle so that the body coordinates of the vehicle conform to the starting point coordinates of the parking route.

Wherein, the several track coordinates have a point coordinate. When the processor module receives a call signal, it can control the positioning unit and the calibration module to start, and confirm the car body Whether the coordinates are the same as the starting point coordinates to confirm whether the vehicle moves from the parking space to the driver's side of the vehicle. If the confirmation result is no, the coordinates of the vehicle body can be obtained from the multiple track coordinates that are closest to the vehicle body. The processor module can send a movement signal to the on-board computer so that the on-board computer can control the movement of the vehicle until the position of the positioning unit The detected vehicle body coordinates are the same as the track coordinates. In this way, it has the function of providing the driver to remotely call the vehicle to a predetermined location.

Wherein, when the parking space is located in an indoor parking lot, the positioning reference object system may be a pillar located in the indoor parking lot, and the correction module may be an ultrasonic unit. When the vehicle moves to the track coordinate The processor module can transmit ultrasonic waves to the pillar through the ultrasonic unit and receive echoes of the ultrasonic waves to obtain the reference distance and reference angle between the vehicle and the pillar. In this way, it has the function of correcting the body coordinates of the vehicle by the position of the pillars in the parking lot.

Wherein, when the parking space is located in an indoor parking lot, the positioning reference object system can be the grid line of the parking space in the indoor parking lot and its parking space number, and the correction module can be an image capturing unit When the vehicle moves to the track coordinate, the processor module can shoot the grid line and its parking space number through the image capturing unit to generate a parking space image. The processor module can be based on the image The processing technology obtains the reference distance and reference angle between the vehicle and the grid line from the parking space image. In this way, it has the function of correcting the vehicle body coordinates of the vehicle with the setting position of the parking space grid line.

The automatic driving assistance system of the present invention may further include a dynamic estimation module electrically connected to the processor module, and has a displacement sensing unit and a yaw angle sensing unit, the displacement sensing unit and the The yaw angle sensing unit is used to detect the displacement and yaw angle of the vehicle moving to the track coordinate to obtain a current displacement change and a yaw angle change. When the vehicle moves to the track coordinate Time, the processor module is based on the current displacement change and the yaw angle change Calculate a parameter value, and calculate the difference between the parameter value and the corresponding preset value of a parameter. If the difference is not equal to zero, a motion compensation signal is sent to the on-board computer to make the on-board computer follow the trajectory The current displacement change and yaw angle change of the coordinates control the vehicle to perform movement compensation. In this way, it can be ensured that the vehicle moves along the several trajectory coordinates, which has the effect of improving the accuracy of parking.

The automatic driving assistance system of the present invention may further include a path generation module that is electrically connected to the processor module. The path generation module can collect an original data when several drivers park their vehicles in the parking space. Parking paths, each of the original parking paths is composed of vehicle body coordinates detected by the positioning unit of each vehicle, and the path generation module uses the several original parking paths as input variables of a type of neural network model to generate The parking path is a driving path that enables the driver on the vehicle to have the best driving comfort during the process of the vehicle moving to the parking space. In this way, the parking route can be continuously improved and updated to the most suitable route through the driving experience-related data of multiple drivers, so that the vehicle can be avoided and in the process of moving along the parking route toward the parking space. The obstacles collide and have a more comfortable driving process, which has the effect of improving parking safety and driving comfort.

Among them, this type of neural network model has another input variable, the other input variable is obtained through a vibration accelerometer and a spectrum analyzer to obtain a vibration acceleration between any two adjacent vehicle body coordinates constituting the original parking path value. In this way, when the vehicle enters the parking space according to the parking path, the vibration generated during the driving of the vehicle can be reduced, which has the effect of further improving driving comfort.

The automatic driving assistance system of the present invention may further include a path generation module electrically connected to the processor module, and the path generation module generates a part of the trajectory coordinates that are closer to the vehicle through the detection capability planning of the correction module , And make the vehicle automatically search for the parking space to compose the parking path according to several track coordinates generated during the vehicle search process. In this way, the system has adaptive Walking effect.

〔this invention〕

1: positioning unit

2: Calibration module

3: processor module

31: wireless transmission unit

4: Dynamic estimation module

41: Displacement sensing unit

42: yaw angle sensing unit

5: Path generation module

C: Vehicle

D: database

E: End point coordinates

L: Positioning reference object

O: starting point

P: parking space

S: Cloud server

[Traditional]

9: Fully automatic parking path determination device

91: Parking Module

92: The first judgment module

93: The second judgment module

94: Third Judgment Module

95: Path Module

96: Control Module

97: positioning unit

[Figure 1] A system block diagram of a conventional fully automatic parking route determination device.

[Figure 2] A system block diagram of a preferred embodiment of the present invention.

[Figure 3] A use case diagram of a preferred embodiment of the present invention.

In order to make the above and other objects, features and advantages of the present invention more comprehensible, the preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings: Please refer to Figure 2. It is a preferred embodiment of the automatic driving assistance system of the present invention. It includes a positioning unit 1, a calibration module 2 and a processor module 3. The positioning unit 1 and the calibration module 2 are provided for installation in a vehicle C, and electrically connect to the processor module 3.

The positioning unit 1 is used to locate the position of the vehicle C to generate a vehicle body coordinate. In this embodiment, the positioning unit 1 can be set at the center of the rear axle of the vehicle C, and can be a real-time dynamic global positioning System (RTK-GPS), a global satellite navigation system (GLONASS) or a Beidou satellite navigation system (COMPASS); or, the positioning unit 1 can also perform positioning by calculating the tire rolling displacement value of the vehicle C in a local area Algorithm to generate the coordinates of the car body, but not limited to this.

The correction module 2 is used to obtain one or more reference distances and reference angles between the vehicle C and a positioning reference object L. The positioning reference object L is used to correct the body coordinates of the vehicle C. The correction module In the second system, a calibration parameter between the vehicle C and the positioning reference object L can be obtained by calculating the reference distance and the reference angle. Preferably, the calibration parameter can be a maximum short distance. In this embodiment, the calibration module 2 may include at least one of an ultrasonic unit and an image capturing unit.

The processor module 3 is electrically connected to the positioning unit 1 and the calibration module 2. The processor module 3 can use a Raspberry Pi 3/4 as a control platform, but is not limited to this. When the processor module 3 receives a parking signal, it controls the positioning unit 1 and the calibration module 2 to activate and receive a parking path, which is composed of several trajectories including a point coordinate O and an end coordinate E It is composed of coordinates, and each of the track coordinates and the positioning reference object L has a relative distance and a relative angle. Specifically, the processor module 3 can be coupled to a cloud server S through a wireless transmission unit 31, and the cloud server S downloads the parking path and the relative relationship between each of the track coordinates and the positioning reference object L. Distance and relative angle; or, the parking path and the relative distances and relative angles can be pre-stored in a database D electrically connected to the processor module 3 for the processor module 3 from the The parking path and the several relative distances and relative angles are obtained from the database D. In this embodiment, the object-oriented programming (OOP) concept can be used to describe several trajectory coordinates of the parking path, and the several trajectory coordinates are stored in the cloud server S or the database D; In addition, the wireless transmission unit 31 can be a wireless transmission module such as 4G, 5G, WiFi, ZigBee, LoRa, Sigfox, NB-IoT or Bluetooth, but it is not limited to this.

When the processor module 3 starts to control the vehicle C to stop to a parking space P along the parking path, that is, before the vehicle C moves along the trajectory coordinates, it can first calculate the vehicle body coordinates and The distance between the trajectory coordinates, and the trajectory coordinate closest to the vehicle body coordinate is taken as the starting point coordinate O of the parking path. The processor module 3 calculates a direction angle and a distance value of the vehicle body coordinate relative to the start coordinate O, and sends a movement signal including the direction angle and the distance value to an on-board computer of the vehicle C (Electronic Control Unit, ECU), so that the on-board computer controls the vehicle C to move to the starting point coordinate O according to the direction angle and the distance value. That Wherein, the processor module 3 and the on-board computer system can be centralized (Centralized), distributed (Distributed) or a combination of the above two. The on-board computer controls the automatic driving technology of the vehicle movement, which belongs to the technical field of the present invention. Those with general knowledge can understand it, so I won’t repeat it here.

The processor module 3 analyzes whether the vehicle body coordinates are the same as the terminal coordinates E to confirm whether the vehicle C is located in the parking space P, and if the analysis result is yes, it means that the vehicle C is already located in the parking space P ; If the result of the analysis is no, then obtain a track coordinate that is closest to the vehicle body coordinate from the several track coordinates, and the track coordinate is located between the vehicle C and the parking space P; At least two consecutive trajectory coordinates that are closest to the vehicle body coordinates are obtained from the trajectory coordinates, and the at least two trajectory coordinates are located between the vehicle C and the parking space P. The processor module 3 sends a movement signal to the on-board computer of the vehicle C so that the on-board computer controls the movement of the vehicle C until the vehicle body coordinates detected by the positioning unit 1 are the same as the track coordinates. The processor module 3 calculates the reference distance and the reference angle detected by the correction module 2 to obtain a correction parameter, and calculates the relative distance and relative angle between the correction parameter and the track coordinates. If the difference is not equal to zero, the processor module 3 sends a motion compensation signal to the on-board computer so that the on-board computer controls the vehicle C to perform motion compensation according to the difference. In addition, when the vehicle C has been parked in the parking space P, the processor module 3 can control the positioning unit 1 and the calibration module 2 to stop operating.

The automatic driving assistance system of the present invention may also have an automatic car-hailing function. Specifically, when the processor module 3 receives a car-hailing signal, it controls the positioning unit 1 and the correction module 2 to start, and confirms the Whether the vehicle body coordinates are the same as the starting point coordinate O to confirm whether the vehicle C moves from the parking space P to the driver of the vehicle C. If the confirmation result is yes, it means that the vehicle C is already located at the driver’s Next to; if the confirmation result is no, the track coordinate closest to the vehicle body coordinate is obtained from the plurality of track coordinates, and the track coordinate is located between the end point coordinate E and the start point coordinate O.

The processor module 3 sends a movement signal to the on-board computer so that the on-board computer controls the movement of the vehicle C until the vehicle body coordinates detected by the positioning unit 1 are the same as the track coordinates. Preferably, the processor module 3 can also control the activation of the calibration module 2 to calculate the calibration parameters between the vehicle C and the positioning reference object L, and calculate the calibration parameters corresponding to the trajectory coordinates If the difference is not equal to zero, the processor module 3 sends a motion compensation signal to the on-board computer so that the on-board computer controls the vehicle C to perform motion compensation according to the difference.

It is worth mentioning that when the automatic driving assistance system of the present invention executes the automatic car-hailing function, the parking position of the vehicle C is not limited to the parking space P, and it can also be located anywhere on both sides of the general road. At this time, the driving A mobile device such as a mobile phone can send a certain point coordinate to the cloud server S, and the cloud server S generates a moving path based on the fixed point coordinate and the location of the vehicle C, and sends the moving path to the processor The module 3 enables the processor module 3 to send a movement signal to the on-board computer, so that the on-board computer controls the vehicle C to move to the driver's side according to the movement path.

The automatic driving assistance system of the present invention may also have a dynamic estimation module 4 electrically connected to the processor module 3. The processor module 3 can determine that the vehicle C moves to the Whether an error occurs in the process of trajectory coordinates, if the judgment result is no, no additional action is required; if the judgment result is yes, a motion compensation signal is sent to the on-board computer so that the on-board computer controls the vehicle according to the error Vehicle C performs movement compensation.

Specifically, the dynamic estimation module 4 may have a displacement sensing unit 41 and a yaw angle sensing unit 42, the displacement sensing unit 41 is used to detect the displacement of the vehicle C to the track coordinate To obtain a current displacement change; the yaw angle sensing unit 42 is used to obtain a yaw angle change when the vehicle C moves to the trajectory coordinate. When the vehicle C moves to the trajectory coordinate, the processor module 3 can be based on the current displacement change and the yaw angle change Calculate a parameter value, and calculate the difference between the parameter value and the corresponding preset value of a parameter. If the difference is not equal to zero, a motion compensation signal is sent to the on-board computer to make the on-board computer follow the track coordinates The current displacement change and yaw angle change control the vehicle to perform movement compensation. Wherein, the parameter value is used to determine whether the vehicle C moves to the trajectory coordinate process and there is an error with the preset value of the parameter. In this embodiment, the parameter value may be a wheel speed value or a tire rolling displacement value. .

The automatic driving compensation system of the present invention may also have a path generation module 5 coupled to the processor module 3, and the path generation module 5 is used to generate the parking path and the movement path. Specifically, when the path generation module 5 is used to generate the parking path, it can collect an original parking path when several drivers park their vehicles C in the parking space P, and each original parking path passes through each The vehicle C is composed of vehicle body coordinates detected by the positioning unit 1. The path generation module 5 uses several original parking paths as input variables of a type of neural network model to generate the parking path. In this embodiment, this type of neural network model can learn a parking path from the several original parking paths. The driver on C has a driving path with the best driving comfort.

Preferably, this type of neural network model can have another input variable, and the other input variable can be obtained through a vibration accelerometer and a spectrum analyzer to obtain the coordinates of any two adjacent car bodies that constitute the original parking path. A vibration acceleration value between. Among them, this type of neural network model can be a single-input single-output model or a multiple-input single-output model, which can be understood by those with ordinary knowledge in the present invention, and will not be repeated here.

In addition, it is noted that the path generation module 5 can also use the detection capability planning of the correction module 2 to generate partial track coordinates that are closer to the vehicle C, and enable the vehicle C to automatically search for the parking space P according to the Several trajectory coordinates generated during the search process of the vehicle C constitute the parking path.

Please also refer to Figure 3, when the driver wants to make his vehicle C drive automatically and park in a parking space P of an indoor parking lot, the positioning reference L can be a pillar located in the indoor parking lot, and The calibration module 2 can be an ultrasonic unit. Specifically, when the processor module 3 receives a parking signal, it controls the positioning unit 1 and the calibration module 2 to start, and can obtain a parking path from the cloud server S or the database D to enable The vehicle C can move along the parking path to one of the parking spaces P of the indoor parking lot. The processor module 3 obtains the vehicle body coordinates of the vehicle C through the positioning unit 1, and analyzes whether the vehicle body coordinates are the same as the terminal coordinates E to confirm whether the vehicle C has been parked in the parking space P, if If the analysis result is negative, a movement signal is sent to the on-board computer of the vehicle C, so that the on-board computer controls the vehicle C to move toward a track coordinate that is closest to the vehicle body coordinate until the positioning unit 1 detects The vehicle body coordinates are the same as the track coordinates.

However, due to the influence of the accuracy of the positioning unit 1, the vehicle body coordinates detected by the positioning unit 1 may have errors with the actual coordinates. Therefore, when the vehicle C moves to the track coordinates, the processing The device module 3 can transmit ultrasonic waves to the pillar through the ultrasonic unit, and receive the echo of the ultrasonic waves, so as to obtain the reference distance and the reference angle between the vehicle C and the pillar, and calibrate the vehicle C accordingly. Body coordinates. The processor module 3 can obtain the relative distance and relative angle between the trajectory coordinates and the pillar from the cloud server S or the database D, and calculate the correction parameters when the vehicle C is located at the trajectory coordinates If the difference is not equal to zero, the processor module 3 sends a motion compensation signal to the on-board computer so that the on-board computer controls the vehicle C to perform motion compensation according to the difference.

On the other hand, when the calibration module 2 is an image capturing unit, the positioning reference object L may be the grid line of the parking space in the indoor parking lot and its parking space number. Specifically, when the vehicle C moves to the track coordinate, the processor module 3 can shoot the grid line and its parking space number through the image capturing unit to generate a parking space image. The processor module 3 According to the image processing technology, the reference distance and the reference angle between the vehicle C and the grid line are obtained from the parking space image, and the trajectory coordinates and the parking position can be obtained from the cloud server S or the database D The relative distance and relative angle between the grid lines of the bit number, and calculate the difference between the correction parameter and the actual value when the vehicle C is located at the trajectory coordinate. If the difference is not equal to zero, the processor module 3 can send A motion compensation signal is sent to the on-board computer, so that the on-board computer controls the vehicle C to perform motion compensation according to the difference.

Those with ordinary knowledge in the present invention can understand that when the positioning reference L is a column in the indoor parking lot, and the grid line of the parking space and its parking space number, the calibration module 2 can also have a super Sound wave unit and an image capture unit. The processor module 3 can first transmit ultrasonic waves through the ultrasonic unit to try to detect the existence of the pillar. If the ultrasonic unit does not detect the pillar, the processor module 3 can The ultrasonic unit is switched to the image capturing unit, and the grid line of the parking space and its parking space number are captured through the image capturing unit to calculate the difference between the correction parameter and the actual value when the vehicle is at the track coordinate . Conversely, the processor module 3 can also use the image capturing unit to try to capture the grid line of the parking space and its parking space number, and when the shooting fails, it can transmit ultrasonic waves through the ultrasonic unit instead. To detect the reference distance and reference angle between the pillar and the vehicle C. In this embodiment, the ultrasonic unit and the image capturing unit can operate independently, switch operation, sequential operation or synchronous operation, which can be understood by those with ordinary knowledge in the relevant field of the present invention.

To sum up, the automatic driving assistance system of the present invention can subdivide the parking path into several trajectory coordinates through a few hardware devices and simple computer processing operations, and make the vehicle only need to move to the bottom during each movement. A track coordinate, and when the vehicle body coordinate detected by the positioning unit is the same as the track coordinate, the processor module obtains the reference distance and the reference distance between the vehicle and the positioning reference object through the calibration module Reference angle to obtain the correction parameter, and calculate the actual value calculated by the relative distance and relative angle between the correction parameter and the trajectory coordinate When the difference is not equal to zero, the on-board computer controls the vehicle to perform movement compensation according to the difference. In this way, the automatic driving assistance system of the present invention has the advantages of simplifying hardware costs, reducing computer computing performance requirements, completing path planning in a guided manner through the several trajectory coordinates, and improving the coordinate accuracy of the positioning unit through the calibration module And other effects.

Although the present invention has been disclosed using the above-mentioned preferred embodiments, it is not intended to limit the present invention. Anyone who is familiar with the art without departing from the spirit and scope of the present invention may make various changes and modifications relative to the above-mentioned embodiments. The technical scope of the invention is protected. Therefore, the scope of protection of the invention shall be subject to the scope of the attached patent application.

1: positioning unit

2: Calibration module

3: processor module

31: wireless transmission unit

4: Dynamic estimation module

41: Displacement sensing unit

42: yaw angle sensing unit

5: Path generation module

D: database

S: Cloud server

Claims (9)

An automatic driving assistance system includes: a positioning unit for installing on a vehicle, the positioning unit for locating the position of the vehicle to generate a vehicle body coordinate; a calibration module for installing on the vehicle, the calibration module The group is used to obtain a reference distance and a reference angle between the vehicle and a positioning reference object; and a processor module, which is electrically connected to the positioning unit and the correction module, and the processor module receives a parking When the signal is signaled, the positioning unit and the calibration module are controlled to activate and receive a parking path. The parking path is composed of several track coordinates including an end point coordinate. Each track coordinate and the positioning reference object have a relative relationship. Distance and a relative angle. The processor module analyzes whether the vehicle body coordinates are the same as the end point coordinates to confirm whether the vehicle is located in a parking space. A track coordinate closest to the vehicle body coordinate, and the track coordinate is located between the vehicle and the parking space, the processor module sends a movement signal to an on-board computer of the vehicle, so that the on-board computer controls the movement of the vehicle , Until the vehicle body coordinates detected by the positioning unit are the same as the track coordinates, the processor module calculates the reference distance and the reference angle detected by the calibration module to obtain a calibration parameter, and calculate the calibration The difference between the relative distance and the relative angle of the parameter corresponding to the trajectory coordinates. If the difference is not equal to zero, a motion compensation signal is sent to the on-board computer so that the on-board computer is based on the difference. Control the vehicle for motion compensation. For example, the automatic driving assistance system of claim 1, wherein the plurality of trajectory coordinates have a point coordinate, and before the vehicle moves along the plurality of trajectory coordinates, the processor module calculates the vehicle body coordinates and each of the trajectory coordinates And take a track coordinate closest to the vehicle body coordinate as the starting point coordinate. The processor module calculates the vehicle body coordinate to obtain a direction angle and a distance value relative to the starting point coordinate, and sends it including A moving signal of the direction angle and the distance value to the on-board electric The brain makes the on-board computer control the vehicle to move to the starting point coordinates according to the direction angle and the distance value. For example, the automatic driving assistance system of claim 1, in which the multiple track coordinates have a point coordinate, when the processor module receives a call signal, it controls the positioning unit and the calibration module to start, and confirms the Whether the vehicle body coordinates are the same as the starting point coordinates to confirm whether the vehicle moves from the parking space to the driver’s side of the vehicle. An approaching track coordinate, and the track coordinate is located between the end point coordinate and the starting point coordinate, the processor module sends a movement signal to the on-board computer so that the on-board computer controls the movement of the vehicle until the positioning unit detects The measured vehicle body coordinates are the same as the track coordinates. For example, the automatic driving assistance system of claim 1, wherein when the parking space is located in an indoor parking lot, the positioning reference object is a pillar located in the indoor parking lot, and the correction module is an ultrasonic unit, When the vehicle moves to the trajectory coordinate, the processor module transmits an ultrasonic wave to the pillar through the ultrasonic unit, and receives the echo of the ultrasonic wave to obtain a reference distance and a reference angle between the vehicle and the pillar. For example, the automatic driving assistance system of claim 1, wherein when the parking space is located in an indoor parking lot, the positioning reference is the grid line of the parking space in the indoor parking lot and its parking space number, and the correction model The group is an image capturing unit. When the vehicle moves to the track coordinate, the processor module shoots the grid line and its parking space number through the image capturing unit to generate a parking space image. The processor The module obtains the reference distance and reference angle between the vehicle and the grid line from the parking space image based on the image processing technology. For example, the automatic driving assistance system of any one of claim 1 to 5 further includes a dynamic estimation module electrically connected to the processor module, and has a displacement sensing unit and a yaw angle sensing unit, the The displacement sensing unit and the yaw angle sensing unit are respectively used to detect the displacement and yaw angle of the vehicle moving to the track coordinate to obtain a current displacement change and a yaw angle change When the vehicle moves to the track coordinate, the processor module calculates a parameter value according to the current displacement change and the yaw angle change, and calculates the parameter value and a corresponding preset value of the parameter If the difference is not equal to zero, a motion compensation signal is sent to the on-board computer so that the on-board computer controls the vehicle to perform motion compensation according to the current displacement change and yaw angle change of the track coordinate. For example, the automatic driving assistance system of any one of claim items 1 to 5 further includes a path generation module electrically connected to the processor module, and the path generation module collects several drivers to park their vehicles in the parking space An original parking path at time, each of the original parking paths is composed of the vehicle body coordinates detected by the positioning unit of each vehicle, and the path generation module uses the several original parking paths as the input of a type of neural network model Variable to generate the parking path, which is a driving path that enables the driver on the vehicle to have the best driving comfort during the process of moving the vehicle to the parking space. For example, the automatic driving assistance system of claim 7, wherein the neural network model of this type has another input variable, and the other input variable is obtained through a vibration accelerometer and a spectrum analyzer to obtain any neighbors that constitute the original parking path A vibration acceleration value between two vehicle body coordinates. For example, the automatic driving assistance system of any one of claim items 1 to 5 further includes a path generation module electrically connected to the processor module, and the path generation module generates closer proximity through the detection capability planning of the calibration module Part of the trajectory coordinates of the vehicle and make the vehicle automatically search for the parking space to form the parking path according to several trajectory coordinates generated during the vehicle search process.

Images