CN106774328A - A kind of automated driving system and method based on road Identification - Google Patents

Abstract

The invention discloses an automatic driving system and method based on road recognition. The system includes a camera, a GPS positioning module, an automatic steering wheel control device and a micro control unit. The camera, the GPS positioning module and an automatic steering wheel control device are respectively It is connected with the micro control unit; there are multiple cameras, and the multiple cameras are placed in various parts of the car body for all-round collection of the surrounding environment information of the car; the GPS positioning module is installed inside the car body for collecting position information; The micro-control unit is used to identify whether the car deviates from the correct route and press the line according to the image information collected by the camera, and control the car through the steering wheel automatic control device according to the position information collected by the GPS positioning module, so that the car returns to Drive on the correct route. The invention can realize the automatic driving of the unmanned vehicle, drive on a specific road without manual intervention, and realize the switching mode of automatic driving and manual driving.

Description

A road recognition-based automatic driving system and method

technical field

The invention relates to an automatic driving system and method, in particular to an automatic driving system and method based on road recognition, belonging to the field of automatic driving of unmanned vehicles.

Background technique

Along with the progress of the society, the living standard of the residents in our country is continuously improved, and the automobile has become an important means of transportation that we cannot live without. However, due to the increasing number of vehicles, the accompanying urban congestion has become a major problem in life and has caused great inconvenience to our travel. At the same time, with the advent of the intelligent era, the Internet of Things has developed rapidly, and the transportation system has also become more intelligent. The self-driving system will greatly reduce urban congestion, and self-driving cars are beneficial to society, drivers and pedestrians. Self-driving cars will effectively reduce human driving errors, the incidence of traffic accidents can be reduced to almost zero, and the cost of traffic congestion can be saved.

Contents of the invention

The purpose of the present invention is to solve the above-mentioned defects in the prior art, and to provide an automatic driving system based on road recognition. Switch mode for driving and manual driving.

Another object of the present invention is to provide an automatic driving method based on road recognition.

The purpose of the present invention can be achieved by taking the following technical solutions:

An automatic driving system based on road recognition, comprising a camera, a GPS positioning module, an automatic steering wheel control device and a micro control unit, wherein the camera, the GPS positioning module and an automatic steering wheel control device are respectively connected to the micro control unit;

There are a plurality of cameras, and the plurality of cameras are placed in various parts of the car body for collecting the surrounding environment information of the car in an all-round way;

The GPS positioning module is installed inside the car body for collecting location information;

The micro-control unit is used to identify whether the car deviates from the correct route and press the line according to the image information collected by the camera, and control the car through the steering wheel automatic control device according to the position information collected by the GPS positioning module, so that the car returns to drive on the correct route;

Further, the micro-control unit has a built-in embedded minimum system, and the embedded minimum system uses the cumulative probability Hough transform algorithm to identify lane lines, perform analysis and processing, and fit out the lane.

Further, the system also includes a real-time control display panel, which is connected with the micro control unit and used to display the driving status of the vehicle in real time.

Further, the real-time control display panel is installed on the right side of the main driver inside the vehicle body.

Further, the automobile steering wheel automatic control device includes a steering wheel, a steering shaft, a geared motor, a worm, a worm gear, an input shaft and an output shaft, and the geared motor is connected to a micro control unit through a CAN bus, and is connected to a worm, and the worm Meshing with the worm gear, the worm is connected to the input shaft, the worm gear is connected to the output shaft, and the output shaft is connected to the steering wheel; after the geared motor rotates, the input shaft drives the worm to rotate, and the worm rotates around the axis. The worm gear is made to rotate, thereby driving the rotation of the output shaft, and the rotation of the output shaft drives the steering wheel to rotate through the steering shaft.

Further, the automobile steering wheel automatic control device also includes a first gear, a second gear and a multi-turn absolute encoder, the axis of the first gear coincides with the axis of the steering wheel, and the first gear and the second gear The two gears are meshed, and the axis of the second gear coincides with the axis of the multi-turn absolute encoder; after the steering wheel rotates, it drives the first gear to rotate, and the rotation of the first gear makes the second gear rotate, thereby Drive the shaft center of the multi-turn absolute encoder to rotate.

Another object of the present invention can be achieved by taking the following technical solutions:

A road recognition-based automatic driving method, the method comprising:

The image information collected by the camera is transmitted to the micro-control unit, and the micro-control unit uses the cumulative probability Hough transform algorithm to identify the lane line based on the image information collected by the camera, conducts analysis and processing, and fits the lane to identify whether the car deviates from the correct route. According to the location information collected by the GPS positioning module, the car is controlled through the steering wheel automatic control device, so that the car can return to the correct route.

Further, the use of the cumulative probability Hough transform algorithm to identify the lane line, perform analysis and processing, and fit the lane, specifically:

Using the cumulative probability Hough transform algorithm to extract the lane line image information from the collected image information, by graying the lane line image and adaptive threshold binarization, the Canny edge detection operator is used to extract the edge information, and according to the lane distribution The features of the lane line predict the position of the lane line in the next frame from the position of the lane line in the previous frame, and finally use the least squares method to fit the road model to match and obtain new lane line information.

Further, the specific steps of the cumulative probability Hough transform algorithm are as follows:

1) Randomly select a pixel point from the point set, and add one to the corresponding accumulator;

2) Delete the point from the point set;

3) Update the accumulator;

4) If the accumulator value after the update is greater than the threshold, then delete all points located on the straight line in the set;

5) Repeat the above steps until the point set is empty.

Further, the adaptive threshold binarization is realized by the method of maximum variance between classes.

Compared with the prior art, the present invention has the following beneficial effects:

1. The present invention adopts a distributed method to place a camera on the vehicle body to observe the surrounding environment in all directions and from multiple angles. It is equipped with an automatic steering wheel control device, and a GPS positioning module is installed inside the vehicle body. The micro-control unit can collect images according to the camera. , Through the effective recognition of the lane markings, it is judged whether there is a lane departure, and according to the position information collected by the GPS positioning module, the steering of the car is controlled by the automatic steering wheel control device to realize automatic steering.

2. The present invention utilizes the cumulative probability Hough transform algorithm to extract lane line image information from the collected image information, by graying the lane line image and adaptive threshold binarization, and using the Canny edge detection operator to extract edge information, The preprocessing of extracting edge information can improve the detection accuracy, and predict the lane line position of the next frame from the position of the lane line in the previous frame according to the characteristics of the lane distribution, so as to improve the efficiency.

3. The present invention installs a real-time control display panel on the right side of the main driver inside the car body, through which the real-time control display panel can display the driving status of the vehicle in real time, and can also provide application functions such as car control, navigation, audio and video.

4. In order to provide users with a more convenient operation mode, the present invention provides a combination of automatic driving and manual driving. The electromagnetic clutch device is used to separate and close the automatic driving device, which can realize one-key switching between automatic driving and manual driving.

5. The present invention utilizes the grayscale features of road boundaries and lane marking lines to complete the identification of road boundaries and lane marking lines, uses CAN bus to communicate with micro-control units, and then controls the steering and accelerator of automobiles to realize unmanned vehicles. purpose of autonomous driving.

Description of drawings

FIG. 1 is a structural block diagram of an automatic driving system based on road recognition according to Embodiment 1 of the present invention.

FIG. 2 is a schematic diagram of the position of the camera according to Embodiment 1 of the present invention.

FIG. 3 is a schematic diagram of the location of the real-time control display panel according to Embodiment 1 of the present invention.

Fig. 4 is a structural diagram of an automatic steering wheel control device for an automobile according to Embodiment 1 of the present invention.

Fig. 5 is a schematic diagram of the connection between the worm and the worm gear in the automatic steering wheel control device of an automobile according to Embodiment 1 of the present invention.

FIG. 6 is a flowchart of an automatic driving method based on road recognition according to Embodiment 2 of the present invention.

Among them, 1-camera, 2-real-time control display panel, 3-GPS positioning module, 4-automatic steering wheel control device, 5-micro control unit, 6-steering wheel, 7-steering shaft, 8-reduction motor, 9-worm , 10-worm gear, 11-input shaft, 12-output shaft, 13-first gear, 14-second gear, 15-multi-turn absolute encoder.

detailed description

The present invention will be further described in detail below in conjunction with the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

Example 1:

As shown in Figure 1, the present embodiment provides a kind of automatic driving system based on road recognition, and this system comprises camera 1, real-time control display panel 2, GPS positioning module 3, automobile steering wheel automatic control device 4 and microcontrol unit (Microcontroller Unit, MCU) 5, the camera 1, the real-time control display panel 2, the GPS positioning module 3, the automobile steering wheel automatic control device 4 are connected to the micro control unit 5 respectively.

As shown in Figure 2, the camera 1 adopts a wide-angle color camera device, and multiple cameras 1 can be placed on the car body. As can be seen from Figure 2, the camera 1 is placed on the front of the car in this embodiment. and the left and right sides, which are used to collect all-round information about the surrounding environment of the car, such as road surfaces and traffic lights.

As shown in FIG. 3 , the real-time control display panel 2 is installed on the right side of the main driver inside the vehicle body, and is used for real-time display of vehicle driving conditions, and can also provide application functions such as vehicle control, navigation, audio and video.

The GPS positioning module 3 is installed inside the vehicle body for collecting position information.

As shown in Fig. 4～Fig. 5, described automobile steering wheel automatic control device 4 comprises steering wheel 6, steering shaft 7, reduction motor 8, worm screw 9, worm wheel 10, input shaft 11, output shaft 12, first gear 13, second Gear 14 and multi-turn absolute value encoder 15;

The geared motor 8 is connected with the micro control unit 5 through the CAN bus, and is connected with the worm 9, and the worm 9 is meshed with the worm wheel 10, and the worm 10 is connected with the input shaft 11, and the worm wheel 10 is connected with the output shaft 12 , the output shaft 12 is connected with the steering wheel; after the reduction motor 8 rotates, the input shaft 11 drives the worm 9 to rotate, and the worm 9 rotates around the axis to make the worm wheel 10 rotate, thereby driving the rotation of the output shaft 12, The rotation of the output shaft 12 drives the steering wheel 6 to rotate through the steering shaft 2;

The shaft center of the first gear 13 coincides with the shaft center of the steering wheel 6, the first gear 13 is meshed with the second gear 14, and the shaft center of the second gear 14 is aligned with the multi-turn absolute value encoder 15. The axes coincide; after the steering wheel 6 rotates, it drives the first gear 13 to rotate, and the rotation of the first gear 13 makes the second gear 14 rotate, thereby driving the axis of the multi-turn absolute value encoder 15 to rotate, and the angular displacement can be obtained The data.

The micro control unit 5 is used to identify whether the car deviates from the correct route and press the line according to the image information collected by the camera 1, and controls the car through the steering wheel automatic control device 4 according to the position information collected by the GPS positioning module 3, Make the car go back to the correct route; Among them, the micro-control unit has an embedded minimum system, and the embedded minimum system uses the cumulative probability Hough transform algorithm to identify the lane line, analyze and process it, and fit the lane.

Example 2:

As shown in Figure 6, this embodiment provides a road recognition-based automatic driving method, which is implemented based on the above-mentioned system, and includes the following steps:

S1, the image information collected by the camera is transmitted to the micro control unit;

S2. According to the image information collected by the camera, the micro control unit uses the cumulative probability Hough transform algorithm (PPHT) to identify the lane line, analyze and process it, and fit the lane, so as to identify whether the car deviates from the correct route and press the line; wherein, the Describes the use of the cumulative probability Hough transform algorithm to identify lane lines, analyze and process them, and fit out lanes, specifically:

Using the cumulative probability Hough transform algorithm to extract the lane line image information from the collected image information, by graying the lane line image and adaptive threshold binarization (using the maximum between-class variance method, also known as the Otsu method OTSU), The Canny edge detection operator is used to extract edge information, and the preprocessing of edge information extraction can improve the detection accuracy, and according to the characteristics of the lane distribution, the position of the lane line in the next frame is predicted from the position of the lane line in the previous frame, that is, the position of the lane line in the previous frame The position of the lane line and boundary in the frame image is used as the predicted position of the next frame to search for new feature points in its horizontal field to improve efficiency. Finally, the least squares method is used to fit the road model to obtain new lane line information .

The cumulative probability Hough transform algorithm is an improvement of the standard Hough transform (SHT) algorithm, which performs Hough transform within a certain range to calculate the direction and range of a single line segment, thereby reducing the amount of calculation and shortening the calculation time; Therefore, PPHT is called "probability" because it does not accumulate all possible points in the accumulator plane, but only accumulates a part of them. If the peak value is high enough, it only takes a small amount of time to find it. Therefore, the calculation time can be substantially reduced, and the specific steps are as follows:

1) Randomly select a pixel point from the point set, and add one to the corresponding accumulator;

2) Delete the point from the point set;

3) Update the accumulator;

4) If the accumulator value after the update is greater than the threshold, then delete all points located on the straight line in the set;

5) Repeat the above steps until the point set is empty.

S3. The micro control unit controls the car through the steering wheel automatic control device 4 according to the position information collected by the GPS positioning module 3, so that the car returns to the correct route.

In summary, the present invention adopts a distributed method to place cameras on the vehicle body to observe the surrounding environment in all directions and from multiple angles. It is equipped with an automatic steering wheel control device, and a GPS positioning module is installed inside the vehicle body. Through the effective recognition of the lane markings in the collected images, it is judged whether there is a lane departure, and according to the position information collected by the GPS positioning module, the steering of the car is controlled by the automatic steering wheel control device to realize automatic steering.

The above is only a preferred embodiment of the patent of the present invention, but the scope of protection of the patent of the present invention is not limited thereto. Equivalent replacements or changes to the technical solutions and their inventive concepts all fall within the scope of protection of the invention patent.

Claims (10)

1. A kind of automatic driving system based on road recognition, it is characterized in that: comprise camera, GPS positioning module, automobile steering wheel automatic control device and micro-control unit, described camera, GPS positioning module and automobile steering wheel automatic control device are respectively connected with micro-controller unit connected; There are a plurality of cameras, and the plurality of cameras are placed in various parts of the car body for collecting the surrounding environment information of the car in an all-round way; The GPS positioning module is installed inside the car body for collecting location information; The micro-control unit is used to identify whether the car deviates from the correct route and press the line according to the image information collected by the camera, and control the car through the steering wheel automatic control device according to the position information collected by the GPS positioning module, so that the car returns to Drive on the correct route. 2. A kind of automatic driving system based on road recognition according to claim 1, characterized in that: said micro-control unit has a built-in embedded minimum system, and the embedded minimum system utilizes the cumulative probability Hough transform algorithm to identify lane lines and perform Analysis and processing, fitting out of the lane. 3. A road recognition-based automatic driving system according to claim 1 or 2, characterized in that: the system also includes a real-time control display panel, the real-time control display panel is connected with a micro-control unit for real-time Displays the driving condition of the vehicle. 4 . The road recognition-based automatic driving system according to claim 3 , wherein the real-time control display panel is installed on the right side of the main driver inside the vehicle body. 5. A kind of automatic driving system based on road recognition according to claim 1 or 2, characterized in that: said automobile steering wheel automatic control device comprises steering wheel, steering shaft, reduction motor, worm screw, worm gear, input shaft and output shaft , the geared motor is connected to the micro control unit through the CAN bus, and is connected to the worm, the worm is meshed with the worm wheel, the worm is connected to the input shaft, the worm wheel is connected to the output shaft, and the output shaft is connected to the steering wheel After the geared motor rotates, the input shaft drives the worm to rotate, and the worm rotates around the axis to make the worm wheel rotate, thereby driving the output shaft to rotate, and the rotation of the output shaft drives the steering wheel to rotate through the steering shaft. 6. A kind of automatic driving system based on road recognition according to claim 5, characterized in that: said automobile steering wheel automatic control device also includes a first gear, a second gear and a multi-turn absolute value encoder, said first The axis of a gear coincides with the axis of the steering wheel, the first gear meshes with the second gear, and the axis of the second gear coincides with the axis of the multi-turn absolute value encoder; the steering wheel rotates Finally, the first gear is driven to rotate, and the rotation of the first gear causes the second gear to rotate, thereby driving the axis of the multi-turn absolute encoder to rotate. 7. An automatic driving method based on road recognition, characterized in that: the method comprises: The image information collected by the camera is transmitted to the micro-control unit, and the micro-control unit uses the cumulative probability Hough transform algorithm to identify the lane line based on the image information collected by the camera, conducts analysis and processing, and fits the lane to identify whether the car deviates from the correct route. According to the location information collected by the GPS positioning module, the car is controlled through the steering wheel automatic control device, so that the car can return to the correct route. 8. The automatic driving method according to claim 7, characterized in that: the use of the cumulative probability Hough transform algorithm to identify lane lines, perform analysis and processing, and fit out the lane, specifically: Using the cumulative probability Hough transform algorithm to extract the lane line image information from the collected image information, by graying the lane line image and adaptive threshold binarization, the Canny edge detection operator is used to extract the edge information, and according to the lane distribution The features of the lane line predict the position of the lane line in the next frame from the position of the lane line in the previous frame, and finally use the least squares method to fit the road model to match and obtain new lane line information. 9. The automatic driving method according to claim 8, characterized in that: the specific steps of the cumulative probability Hough transform algorithm are as follows: 1) Randomly select a pixel point from the point set, and add one to the corresponding accumulator; 2) Delete the point from the point set; 3) Update the accumulator; 4) If the accumulator value after the update is greater than the threshold, then delete all points located on the straight line in the set; 5) Repeat the above steps until the point set is empty. 10. The automatic driving method according to any one of claims 7-9, characterized in that: said adaptive threshold binarization adopts a maximum inter-class variance method.