A lane keeping system adjustment method based on driver's psychological characteristics
Through deep learning neural networks, identify large vehicles and calculate safe distances, adjust the autonomous driving lane keeping system, solve the problem that autonomous driving cannot avoid large vehicles, reduce the psychological burden and insufficient reaction time of drivers, and improve the intelligence of autonomous driving level.
Patent Information
- Application Number
- CN202210883113.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-26
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The autonomous lane keeping system cannot actively avoid it like a human driver when encountering large vehicles, which increases the driver's psychological burden and insufficient reaction time.
Deep learning neural network is used to identify large vehicles in the road environment, establish and expand the data set during overtaking, calculate safe spacing, and adjust the spacing between vehicles and lane lines according to the lane maintenance system to achieve similar avoidance to human drivers measure.
Under autonomous driving, the same avoidance measures as those of human drivers are achieved, reducing the psychological burden on drivers, leaving drivers with more reaction time, and making autonomous driving more intelligent.
Smart Images

Figure CN115214657B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of autonomous driving safety technology, and in particular to an adjustment method for a lane keeping system based on driver psychological characteristics. Background Art
[0002] In real life, when a driver of a small vehicle encounters a large vehicle driving in the right lane in the same direction, the driver of the small car will generally take the initiative to avoid the large vehicle when overtaking in the left lane, try to overtake on the left side of the lane, and keep a sufficient lateral distance from the large vehicle as much as possible. This behavior of the driver of the small car can not only make themselves feel more psychologically secure, but also give the driver more reaction time to avoid the large vehicle in time when an abnormal emergency occurs.
[0003] At present, the lane keeping system of autonomous driving will automatically maintain the position of the vehicle in the lane, automatically obtain the positional relationship between the vehicle and the lane line based on the visual sensor, and automatically adjust the steering wheel angle in real time to ensure that the vehicle is always in the middle of the lane and does not deviate. This is both the advantage and the defect of the lane keeping system. Therefore, when overtaking a large vehicle, the lane keeping system will not actively avoid the large vehicle like the driver of a small car, but will only keep the vehicle in the middle of the lane, thereby increasing the psychological burden of the driver of the small car and giving the driver less time to react to sudden situations of the large vehicle. Summary of the invention
[0004] In view of the problems existing in the prior art, the object of the present invention is to provide an adjustment method of a lane keeping system based on the psychological characteristics of the driver.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions to achieve it.
[0006] A method for adjusting a lane keeping system based on driver psychological characteristics comprises the following steps:
[0007] Step 1: The camera collects the road environment image around the vehicle, and uses the deep learning neural network to identify whether there is a large vehicle in front of the right side of the vehicle in the road environment image;
[0008] Step 2: calibrate the left lane line of the lane where the large vehicle is located in the road environment image, and identify the distance between the vehicle and the calibrated lane line and the distance between the large vehicle and the calibrated lane line;
[0009] Step 3, establishing a data set of driving parameters during the overtaking process;
[0010] Step 4, expand the data set and establish the relevant function according to the expanded data set;
[0011] Step 5: When a large vehicle is recognized again, the safe distance between the vehicle and the marked lane line is obtained according to the relevant functions and parameters;
[0012] Step 6: Output the safe distance to the lane keeping system according to whether the vehicle has the conditions to change lanes to the left during the process of overtaking the large vehicle;
[0013] Step 7: The lane keeping system adjusts the driving state of the vehicle according to the output safety distance, so that the distance between the vehicle and the calibrated lane line is not less than the output safety distance.
[0014] Compared with the prior art, the beneficial effect of the present invention is that when the self-driving vehicle encounters a large vehicle, the self-driving vehicle can implement the same avoidance measures as the driver, thereby meeting the driver's psychological demands, reducing the driver's psychological burden, leaving the driver with more reaction time, and making the self-driving vehicle more intelligent. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0016] Figure 1 The present invention is a flow chart of a method for adjusting a lane keeping system based on driver psychological characteristics. DETAILED DESCRIPTION
[0017] The embodiments of the present invention will be described in detail below with reference to examples, but those skilled in the art will understand that the following examples are only used to illustrate the present invention and should not be construed as limiting the scope of the present invention.
[0018] A method for adjusting a lane keeping system based on driver psychological characteristics comprises the following steps:
[0019] Step 1: The camera collects the road environment image around the vehicle, and uses the deep learning neural network to identify whether there is a large vehicle in front of the right side of the vehicle in the road environment image;
[0020] Specifically, the deep learning YOLO V3 network model algorithm is used to perform image semantic segmentation on the road environment image to identify whether there is a vehicle in the right front. If there is a vehicle, the width of the vehicle is identified. When the width of the vehicle is greater than 2.2m, the vehicle is considered to be a large vehicle.
[0021] Step 2: calibrate the left lane line of the lane where the large vehicle is located in the road environment image, and identify the distance between the vehicle and the calibrated lane line and the distance between the large vehicle and the calibrated lane line;
[0022] The distance between the vehicle and the calibrated lane line, as well as the distance between a large vehicle and the calibrated lane line, can be obtained through the lane line sensor and the road environment image after the lane line is calibrated.
[0023] Step 3, establishing a data set of driving parameters during the overtaking process;
[0024] Specifically, when the maximum distance between the ego vehicle and the calibrated lane line is greater than or equal to the lane width during the overtaking process, it means that the ego vehicle changed lanes to the left during the overtaking process. The maximum distance between the ego vehicle and the calibrated lane line, the minimum distance between the large vehicle and the calibrated lane line, and the width of the large vehicle during the overtaking process are recorded as a set of data in data set A.
[0025] When the maximum distance between the ego vehicle and the calibrated lane line is less than the lane width during the overtaking process, it means that the ego vehicle did not change lanes when overtaking. The maximum distance between the ego vehicle and the calibrated lane line, the minimum distance between the large vehicle and the calibrated lane line, and the width of the large vehicle during the overtaking process are recorded as a set of data in data set B.
[0026] Step 4, expand the data set and establish the relevant function according to the expanded data set;
[0027] Specifically, each time the ego vehicle overtakes a large vehicle, the dataset A or dataset B is expanded and updated;
[0028] After the ego vehicle has overtaken the large vehicle N times, a functional relationship between the distance between the ego vehicle and the calibrated lane line, the distance between the large vehicle and the calibrated lane line, and the width of the large vehicle is established based on the data set A, which is recorded as the variable lane function;
[0029] After the ego vehicle has overtaken the large vehicle N times, a functional relationship among the distance between the ego vehicle and the calibrated lane line, the distance between the large vehicle and the calibrated lane line, and the width of the large vehicle is established based on data set B, which is recorded as the immutable lane function.
[0030] Step 5: When a large vehicle is recognized again, the safe distance between the vehicle and the marked lane line is obtained according to the relevant functions and parameters;
[0031] Specifically, when a large vehicle is recognized again, the distance between the large vehicle and the marked lane line and the width of the large vehicle are substituted into the variable lane function and the immutable lane function respectively to obtain the safety distance D 1 And safety distance D 2 .
[0032] Step 6: Output the safe distance to the lane keeping system according to whether the vehicle has the conditions to change lanes to the left during the process of overtaking the large vehicle;
[0033] Specifically, when the vehicle is overtaking a large vehicle and there is no condition to change lanes to the left, the safety distance D is output. 2 ;
[0034] When the vehicle is overtaking a large vehicle and has the conditions to change lanes to the left, determine the safe distance D between the vehicle and the marked lane line. 2 Is it less than the minimum value of the maximum distance between the vehicle and the calibrated lane line in data set B or greater than the maximum value of the maximum distance between the vehicle and the calibrated lane line in data set B? If yes, output the safe distance D 1 ; If not, output the safety distance D 2 .
[0035] For example, when the road is a two-lane road in the same direction or a two-lane road in both directions, when the vehicle overtakes a large vehicle, it can only drive in the left lane to complete the overtaking, and there is no condition to change lanes to the left during the overtaking process; when the road is a three-lane road in the same direction and the large vehicle is in the rightmost lane, when the vehicle overtakes the large vehicle, it can drive in the middle lane or the leftmost lane to complete the overtaking, and at this time the vehicle has the conditions to change lanes to the left during the overtaking process.
[0036] Step 7: The lane keeping system adjusts the driving state of the vehicle according to the output safety distance, so that the distance between the vehicle and the calibrated lane line is not less than the output safety distance.
[0037] Although the present invention has been described in detail in general terms and in specific embodiments in this specification, it is obvious to those skilled in the art that some modifications or improvements may be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection claimed by the present invention.
Claims
1. A method for adjusting a lane keeping system based on driver psychological characteristics, characterized in that: The following steps are involved: Step 1: The camera collects the road environment image around the vehicle, and uses the deep learning neural network to identify whether there is a large vehicle in front of the right side of the vehicle in the road environment image; Step 2: calibrate the left lane line of the lane where the large vehicle is located in the road environment image, and identify the distance between the vehicle and the calibrated lane line and the distance between the large vehicle and the calibrated lane line; Step 3, establish a data set of driving parameters during the overtaking process; when the maximum distance between the vehicle and the calibrated lane line is greater than or equal to the lane width during the overtaking process, the maximum distance between the vehicle and the calibrated lane line, the minimum distance between the large vehicle and the calibrated lane line, and the width of the large vehicle during the overtaking process are recorded as a set of data in data set A; When the maximum distance between the vehicle and the marked lane line is less than the lane width during the overtaking process, the maximum distance between the vehicle and the marked lane line, the minimum distance between the large vehicle and the marked lane line, and the width of the large vehicle during the overtaking process are recorded as a set of data in data set B; Step 4, expand the data set, and establish a related function based on the expanded data set; each time the ego vehicle overtakes a large vehicle, expand and update the data set A or data set B; when the ego vehicle overtakes the large vehicle N times, establish a functional relationship between the distance between the ego vehicle and the calibrated lane line, the distance between the large vehicle and the calibrated lane line, and the width of the large vehicle based on data set A, which is recorded as a variable lane function; when the ego vehicle overtakes the large vehicle N times, establish a functional relationship between the distance between the ego vehicle and the calibrated lane line, the distance between the large vehicle and the calibrated lane line, and the width of the large vehicle based on data set B, which is recorded as an immutable lane function; Step 5: When a large vehicle is recognized again, the safe distance between the vehicle and the marked lane line is obtained according to the relevant functions and parameters; When a large vehicle is recognized again, the distance between the large vehicle and the marked lane line and the width of the large vehicle are substituted into the variable lane function and the immutable lane function respectively to obtain the safety distance D1 and the safety distance D2 respectively; Step 6: Output the safe distance to the lane keeping system according to whether the vehicle has the conditions to change lanes to the left during the process of overtaking the large vehicle; When the vehicle overtakes a large vehicle and there is no condition to change lanes to the left, the safety distance D2 is output; When the vehicle is overtaking a large vehicle and has the conditions to change lanes to the left, determine whether the safety distance D2 between the vehicle and the calibrated lane line is less than the minimum value of the maximum distance between the vehicle and the calibrated lane line in data set B or greater than the maximum value of the maximum distance between the vehicle and the calibrated lane line in data set B; if yes, output the safety distance D1; if no, output the safety distance D2; Step 7: The lane keeping system adjusts the driving state of the vehicle according to the output safety distance, so that the distance between the vehicle and the calibrated lane line is not less than the output safety distance.
2. The lane keeping system adjustment method based on driver psychological characteristics according to claim 1, characterized in that: Step 1 Specifically, the deep learning YOLO V3 network model algorithm is used to perform image semantic segmentation on the road environment image to identify whether there is a vehicle in the right front. If there is a vehicle, the width of the vehicle is identified. When the width of the vehicle is greater than 2.2m, the vehicle is considered to be a large vehicle.
