Steering intention decision-making method considering driving habits of driver with exercise intensity

Abstract

The invention is suitable for the technical field of automobile driving, and provides a steering intention decision-making method considering the exercise intensity driving habit of a driver, and the method comprises the steps: building a probability model of an operation habit and the exercise intensity habit through the steering driving data of the driver under an ideal driving condition, and building the relation between the two; according to the actual operation input of the driver, the exercise intensity conforming to the habits of the driver at the moment, namely the expected lateral acceleration, can be obtained. As the data for constructing the model comes from the driving data of the target driver under the ideal driving condition, the natural driving habit of the current driver can be well reflected, so that the driver can feel more comfortable under the exercise intensity, and a foundation is laid for designing a personalized intelligent driving control strategy of the driver.

Description

Technical Field

[0001] The invention belongs to the technical field of automobile driving, and in particular relates to a steering intention decision method taking into account the driving habit of a driver's sports intensity. Background Art

[0002] In recent years, with the rapid development and change of the automobile industry, the driving and riding experience of automobiles has been greatly improved, and along with it, consumers have more comprehensive expectations for automobile use. Therefore, when purchasing a car, consumers are no longer only concerned with meeting the basic needs of daily transportation, but are further considering whether the driving experience meets their own psychological expectations. Therefore, domestic and foreign practitioners have conducted many studies to consider the impact of the driver's driving style when designing a car, and provide different driving experiences for drivers with different styles to improve consumers' driving satisfaction.

[0003] At present, there are generally the following problems in the driver's driving style recognition method and the design of personalized exercise intensity control strategies for drivers with different driving styles:

[0004] (1) The current driver classification method used by the industry seeks to divide a group of people into one category. However, for some individuals with fuzzy driving styles between the two, there is a possibility that this type of people may not be adapted to the control strategy corresponding to their classification results.

[0005] (2) There are also differences between drivers in the same category. Due to the differences between individual drivers, the preferred driving experience in the same category will also be different. Therefore, for drivers of the same style, using the same sports intensity control strategy cannot well meet the personalized needs of each driver.

[0006] For a single driver, by collecting the characteristics of his driving operation habits, analyzing the motion intensity under his steering condition and making relevant mappings, it is more in line with the driver's motion intensity expectations and more in line with his desired personalized experience. Therefore, in order to solve the above problems, a steering intention decision method considering the driver's motion intensity driving habits is proposed. Summary of the invention

[0007] The purpose of the embodiments of the present invention is to provide a steering intention decision method that takes into account the driver's driving habits of sports intensity, aiming to solve the problems raised in the above-mentioned background technology.

[0008] The embodiment of the present invention is implemented as follows: a steering intention decision method considering the driver's driving habit of sports intensity includes the following steps:

[0009] Step 1: Under ideal driving conditions, the steering wheel torque information of the driver during steering driving is collected, and the steering wheel torque is used to represent the driver's steering operation input. The probability model of driving input habits is constructed using normal distribution. The mean and standard deviation estimators are calculated based on sample data using moment estimation method to complete the parameter analysis of the model, and finally the probability model of operation habits is obtained;

[0010] Step 2: Under ideal driving conditions, the driver's lateral acceleration data during steering is collected, and the lateral acceleration is used to characterize the intensity of the steering movement. The probability model of the intensity of the movement habit is constructed using the normal distribution. Based on the sample data, the corresponding mean and standard deviation estimators are calculated using the moment estimation method to complete the parameter analysis of the model, and then the probability model of the driving intensity of the movement habit is obtained.

[0011] Step 3: Take the characteristic points of the normal distribution curve in the driving operation input habit probability model and establish a corresponding relationship with the characteristic points in the normal distribution curve of the lateral acceleration motion intensity probability model during driving. Since the steering operation is relatively complex and the frequency of use of uncommon operation inputs needs to be considered, the lateral acceleration corresponding to the rollover threshold of the car in the general state is taken as a y =0.7g is used as the boundary point to limit the fitting range of the curve, and an exponential function is used to fit the mapping curve; thus, the lateral movement intensity that the driver is more inclined to in his driving habits can be obtained according to the steering operation input of the driver.

[0012] Step 4: Obtain the steering wheel torque T when turning right sw_run , and the fitting characteristic points of lateral acceleration (2μ1-T sw_lim , 2μ2-a y_lim )、(T sw_lim , a y_lim ) and (T sw_max , 0.7g), a fitting relationship between the two is constructed based on the psychophysical method, and the function when turning left is supplemented into a complete odd function to cover all working conditions of turning left and right; according to the mapping relationship between the two, the corresponding lateral acceleration can be analyzed through the steering torque input of the driver's steering wheel in the current state.

[0013] Another object of an embodiment of the present invention is to provide a steering intention decision module that considers the driver's sports intensity driving habits, based on the above-mentioned steering intention decision method that considers the driver's sports intensity driving habits, including a probability model building module, an operation input information acquisition module and a steering driving intention analysis module;

[0014] The probability model building module is used to build a driver's steering wheel steering torque operation habit probability model and a sports intensity driving habit probability model;

[0015] The operation input information acquisition module is used to collect steering wheel steering torque information and send the data to the steering driving intention analysis module;

[0016] The steering driving intention analysis module is used to analyze the corresponding lateral acceleration of the driver according to the current steering wheel steering torque.

[0017] The embodiment of the present invention provides a steering intention decision method that takes into account the driver's driving habit of sports intensity. The method constructs a probability model of operation habit and sports intensity habit through the driver's steering driving data under ideal driving conditions, and establishes a connection between the two. According to the driver's actual operation input, the sports intensity that meets the driver's habits at this time, that is, the expected lateral acceleration, can be obtained. Since the data for constructing the model comes from the target driver's driving data under ideal driving conditions, it can better reflect the current driver's natural driving habits, so the driver will feel more comfortable under this sports intensity, which lays a foundation for designing the driver's personalized intelligent driving control strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the system principle;

[0019] Figure 2 Build a flowchart for the probabilistic model;

[0020] Figure 3 Fit the curve to the map. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0022] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0023] An embodiment of the present invention provides a steering intention decision method considering a driver's driving habit of sports intensity, comprising the following steps:

[0024] Step 1: Under ideal driving conditions, collect the driver's steering driving related data. Use normal distribution to build a probability model of driving input habits, and select the steering torque of the steering wheel as the driver's steering driving operation input. Since steering is divided into left steering and right steering, the collected steering wheel steering torque T of left steering is sw Calibrated as a positive value, assuming that the sample data of the steering wheel collected in the experiment is N, which conforms to the following normal distribution:

[0025]

[0026] First, using the sample data T sw_i Calculate the sample mean and sample variance S 2 :

[0027]

[0028] Then, using the moment estimation method, we establish an equation based on the population moment being equal to the sample moment:

[0029]

[0030] Combining the expectation and variance formulas of the normal distribution:

[0031] E(T sw )=μ1

[0032] μ1 and The moment estimator of and

[0033]

[0034] Substituting the estimated values ​​of the mean and standard deviation, we obtain the probability model of the driver's steering input operation habits;

[0035] Step 2: Use lateral acceleration to characterize the intensity of motion. According to the lateral acceleration data collected under ideal driving conditions, use the same method to process and obtain the lateral acceleration a. y Probabilistic model of exercise intensity habits:

[0036]

[0037] Step 3: The driving operation input habit probability model reflects the probability distribution of the steering wheel torque applied by the driver. The larger the probability density corresponding to the steering wheel torque value, the more often the driver uses the corresponding steering wheel torque for operation; conversely, it indicates that the driver rarely applies the corresponding steering wheel torque, which well describes the driver's steering operation habits. Similarly, the probability model of motion intensity reflects the probability of the driver selecting different lateral accelerations during driving, which well describes the driver's motion intensity habits when steering.

[0038] After the two probability models are constructed, a mapping operation is performed between the two. The characteristic points of the normal distribution curve in the driving operation input habit probability model are taken to establish a corresponding relationship with the characteristic points in the normal distribution curve of the lateral acceleration motion intensity probability model during driving. Since the steering operation is relatively complex and the frequency of use of uncommon operation inputs needs to be considered, the lateral acceleration corresponding to the rollover threshold of the car in the general state is taken as a y =0.7g is used as the boundary point to limit the fitting range of the curve, and an exponential function is used to fit the mapping curve. Thus, the lateral motion intensity that the driver tends to prefer in his driving habits can be obtained according to the steering operation input of the driver;

[0039] Step 4: According to the psychophysical method, the feature point fitting curve is used to realize the mapping between operation habits and exercise intensity habits, as follows:

[0040] First, the driver's steering wheel torque is mapped to the vehicle's lateral acceleration data. This involves the knowledge of psychophysics. Psychophysics is a branch of psychology that studies the quantitative relationship (functional relationship) between "psychological quantities" and "physical quantities". It establishes a connection between human feelings and experimentally measured data, and describes the connection between complex psychological feelings and external stimuli through a concise and clear mathematical model, thereby exploring the mathematical laws of human feelings when they are experiencing external stimuli.

[0041] Therefore, based on the psychophysical method, the normal distribution center means μ1 and μ2 of the driving operation input habit probability model and the lateral acceleration motion intensity habit probability model are selected as the first pair of mutually mapped feature points (μ1, μ2).

[0042] Since the mapping is based on the driver's habits, the low-probability events at both ends may correspond to emergency or special situations, which are not suitable for characterizing the driver's habits. Therefore, most of the center of the normal distribution is selected for mapping, and 90% of the normal distribution area centered on the mean μ1 and μ2 is selected for mapping, that is:

[0043] P(|T sw -μ1|≤T sw_lim )=0.90

[0044] P(|a y -μ2|≤a y_lim )=0.90

[0045] 2μ1-T sw_lim ≤T sw ≤T sw_lim

[0046] 2μ2-a y_lim ≤a y ≤ay_lim

[0047] Respectively (2μ1-T sw_lim , 2μ2-a y_lim ), (T sw_lim , a y_lim ) as the characteristic point of the steering wheel steering torque and steering driving motion habits within the range of the driver's common driving habits;

[0048] At the same time, the lateral acceleration threshold (considered to be a) of the vehicle driven by the driver near the rollover threshold is measured. y =0.7g) corresponding to the steering wheel steering torque T sw_max , as the boundary critical feature point (T sw_max , 0.7g).

[0049] Assume that the steering wheel steering torque to be mapped is T sw_run , first according to the algebraic relationship of the exponential function:

[0050]

[0051] Substitute the three feature points into the above algebraic relationship to fit the mapped motion intensity function relationship for calibration:

[0052]

[0053] Since the steering wheel steering torque and lateral acceleration information when turning left were used in the early data processing, and the former was calibrated as a positive value, the mapping relationship for right turning was missing. In this regard, we simply supplemented the function when turning left into a complete odd function, and clarified that the steering wheel steering torque when turning right is a negative value, that is, the functional relationship mapping under all steering torques, and still fitted the above mapping function relationship, and merged it into the final complete mapping function relationship.

[0054] Another embodiment of the present invention provides a steering intention decision module considering the driver's driving habits of sports intensity, based on the above-mentioned steering intention decision method considering the driver's driving habits of sports intensity, including a probability model building module, an operation input information collection module and a steering driving intention analysis module;

[0055] The probability model building module is used to build a driver's steering wheel steering torque operation habit probability model and a sports intensity driving habit probability model;

[0056] The operation input information acquisition module is used to collect steering wheel steering torque information and send the data to the steering driving intention analysis module;

[0057] The steering driving intention analysis module is used to analyze the corresponding lateral acceleration of the driver according to the current steering wheel steering torque.

[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A steering intention decision method considering the driver's driving habit of sports intensity, characterized in that: The following steps are involved: Step 1: Under ideal driving conditions, the steering wheel torque information of the driver during steering driving is collected, and the steering wheel torque is used to represent the driver's steering operation input. The probability model of driving input habits is constructed using normal distribution. The mean and standard deviation estimators are calculated based on sample data using moment estimation method to complete the parameter analysis of the model, and finally the probability model of operation habits is obtained; Step 2: Under ideal driving conditions, collect the driver's lateral acceleration data during steering, use the lateral acceleration to characterize the intensity of the steering movement, and use the normal distribution to build a probability model of the intensity of movement habits; Based on the sample data, the corresponding mean and standard deviation estimators are calculated using the moment estimation method to complete the parameter analysis of the model, and then the probability model driving exercise intensity habits is obtained; Step 3: Take the characteristic points of the normal distribution curve in the driving operation input habit probability model and the characteristic points in the normal distribution curve of the lateral acceleration motion intensity probability model during driving to establish a corresponding relationship, and take the lateral acceleration corresponding to the rollover threshold of the car under normal conditions as a y = 0.7g is used as the boundary point to limit the fitting range of the curve, and an exponential function is used to fit the mapping curve; thus, the lateral movement intensity that the driver is more inclined to in his driving habits can be obtained according to the steering operation input of the driver; Step 4: Obtain the steering wheel torque T when turning right sw_run , and the fitting characteristic points of lateral acceleration (2μ1-T sw_lim , 2μ2-a y_lim )、(T sw_lim , a y_lim ) and (T sw_max , 0.7g), a fitting relationship between the two is constructed based on the psychophysical method, and the function when turning left is supplemented into a complete odd function to cover all working conditions of turning left and right; according to the mapping relationship between the two, the corresponding lateral acceleration can be analyzed through the steering torque input of the driver's steering wheel in the current state.

2. The steering intention decision method considering the driver's sports intensity driving habits according to claim 1 is characterized in that: In step 1, the construction of the driver's steering input operation habit probability model includes the following specific steps: The probability model of driving input habits is constructed using normal distribution. The steering torque of the steering wheel is selected as the driver's steering driving operation input. Since steering is divided into left steering and right steering, the collected steering torque of the left steering wheel T sw Calibrated as a positive value, assuming that the sample data of the steering wheel collected in the experiment is N, which conforms to the following normal distribution: First, using the sample data t sw_i Calculate the sample mean and sample variance S 2 : Then, using the moment estimation method, we establish an equation based on the population moment being equal to the sample moment: Combining the expectation and variance formulas of the normal distribution: E(T sw )=μ1 μ1 and The moment estimator of and Substituting the estimated values ​​of the mean and standard deviation, we obtain the probability model of the driver's steering input operation habits.

3. The steering intention decision method considering the driver's driving habit of sports intensity according to claim 2 is characterized in that: In step 2, lateral acceleration is selected to represent the intensity of movement. According to the lateral acceleration data collected under ideal driving conditions, the data is processed in the same manner as step 1 to obtain a probability model of lateral acceleration intensity habit:

4. The steering intention decision method considering the driver's driving habit of sports intensity according to claim 3 is characterized in that: The step 4 comprises the following specific steps: First, the mapping relationship between the driver's steering wheel torque and the vehicle's lateral acceleration data is established. Based on the psychophysical method, the normal distribution center means μ1 and μ2 of the driving operation input habit probability model and the lateral acceleration motion intensity habit probability model are selected as the first pair of mutually mapped feature points (μ1, μ2). The 90% normal distribution area centered on the means μ1 and μ2 is selected for mapping operation, namely: P(|T sw -μ1|≤T sw_lim )=0.90 P(|a y -μ2|≤a y_lim )=0.90 2μ1-T sw_lim ≤T sw ≤T sw_lim 2m2-a y_lim ≤a y ≤a y_lim Respectively (2μ1-T sw_lim , 2μ2-a y_lim ), (T sw_lim , a y_lim ) as the characteristic point of the steering wheel steering torque and steering driving motion habits within the range of the driver's common driving habits; At the same time, the steering wheel steering torque T corresponding to the lateral acceleration threshold near the rollover threshold of the vehicle driven by the driver is measured. sw_max , as the boundary critical feature point (T sw_max , 0.7 g); Assume that the steering wheel steering torque to be mapped is T sw_run , first according to the algebraic relationship of the exponential function: Substitute the three feature points into the above algebraic relationship to fit the mapped motion intensity function relationship for calibration: The function for turning left is supplemented into a complete odd function, and it is clear that the steering torque of the right-turn steering wheel is a negative value, that is, the functional relationship mapping under all steering torques is still fitted by the above mapping function relationship and merged into the final complete mapping function relationship.

5. A steering intention decision module considering the driver's driving habits based on the steering intention decision method considering the driver's driving habits based on any one of claims 1 to 4, characterized in that: It includes a probability model building module, an operation input information collection module and a steering driving intention analysis module; The probability model building module is used to build a probability model of the driver's steering wheel steering torque operation habit and a probability model of the driving intensity driving habit; The operation input information acquisition module is used to collect steering wheel steering torque information and send the data to the steering driving intention analysis module; The steering driving intention analysis module is used to analyze the corresponding lateral acceleration of the driver according to the current steering wheel steering torque.

Images