Automatic driving takeover method and system based on road impedance in intelligent connected environment

Abstract

The present invention belongs to the field of road vehicle control systems and relates to a method and system for automatic driving takeover based on road impedance in an intelligent network environment. The system includes a following model, an impedance value calculation model, a takeover interval determination module, and a human-computer interaction module. The following model is used to determine the movement mode of the vehicle in different road environments. The impedance value calculation model is used to quantify the interference of the road environment on the automatic driving vehicle and convert the complex road environment into a calculable impedance value. The takeover interval determination module determines the speed interval for manual takeover by analyzing the rate of change of the impedance value with the average speed of the vehicle on the road. The human-computer interaction module is used to remind the driver to take over when the speed reaches the threshold of the speed interval for manual takeover. The system can update the speed interval for manual takeover in real time according to the road environment, and promptly remind the driver to perform the takeover operation at the takeover node, thereby improving the driver's user experience and the reliability of the automatic driving vehicle.

Description

Technical Field

[0001] The present invention belongs to the field of road vehicle control systems and relates to autonomous driving vehicle takeover, and specifically to an autonomous driving takeover method and system based on road impedance in an intelligent network environment. Background Art

[0002] Autonomous driving technology is still at Level 3 (conditional autonomous driving), which still requires driver takeover. Takeover refers to the process where a human driver temporarily assumes control of the vehicle when the vehicle encounters road conditions it cannot handle. To improve the reliability of autonomous vehicles and the driver's user experience, a takeover system is needed to determine the speed range within which the driver can take over.

[0003] Traditional takeover systems set the corresponding steering wheel torque or judge the takeover performance according to different road environments to ensure the stable operation of autonomous vehicles. This method has the following disadvantages and shortcomings: (1) Lack of real-time performance. When traffic conditions suddenly change, the torque cannot be adjusted in advance; (2) Delayed response. Judging the driver's status by taking over performance may miss the best time to take over; (3) Lack of real-time feedback. After confirming manual takeover, the lack of human-computer interaction hardware (tactile feedback, visual warnings and sound prompts) will affect the takeover efficiency.

[0004] In an intelligent connected environment, autonomous vehicles can monitor the road environment in real time and obtain the operating conditions of surrounding vehicles. This not only improves perception and decision-making capabilities but also provides a foundation for the vehicle's driving strategy. Therefore, it is necessary to provide an autonomous driving takeover system that can receive real-time road information data. Summary of the Invention

[0005] In view of the above-mentioned technical problems and defects, the purpose of the present invention is to provide an automatic driving takeover method based on road impedance in an intelligent connected environment. The takeover method calculates the impedance value of the road through real-time acquired road environment data, and determines the operating efficiency of the automatic driving vehicle under different road environments based on the impedance value, thereby determining the manual takeover speed range, improving the reliability of automatic driving takeover and the driver's user experience, and providing technical support for the development of intelligent transportation systems.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for autonomous driving takeover based on road impedance in an intelligent connected environment, the method comprising the following steps:

[0008] Step 1. Construct a car-following model to determine the motion patterns of manually driven vehicles and autonomous vehicles under different road conditions. The car-following model includes a car-following model for manually driven vehicles and a car-following model for autonomous vehicles. The car-following model for manually driven vehicles is used to represent the motion patterns of manually driven vehicles, and is expressed as:

[0009] a n (t) = λv n (t) m {v n-1 (t)-v n (t)+δ[x n-1 (t)-x n (t)-X]};

[0010] X=kv n (t)+b;

[0011] Where a n (t) is the acceleration of the following vehicle at the current moment, λ is the sensitivity coefficient, v n (t) is the speed of the following vehicle at the current moment, m is the coefficient to be determined, v n-1 (t) is the speed of the preceding vehicle at the current moment, δ is the weight coefficient of the distance term, and x n-1 (t) is the position of the preceding vehicle at the current moment, x n (t) is the position of the following vehicle at the current moment, X is the expected following distance of the following vehicle, k and b are unknown coefficients;

[0012] The car-following model of the autonomous driving vehicle is used to represent the movement mode of the autonomous driving vehicle;

[0013] Step 2. Determine the average speed of vehicles on the road based on the current road environment, and determine the vehicle density of the road based on the car-following model for manually driven vehicles and the car-following model for autonomous vehicles.

[0014]

[0015] Where, t av is the expected headway of the autonomous vehicle, L is the length of the autonomous vehicle, S0 is the minimum safe distance, m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′+n=1, and V is the average speed of vehicles on the road;

[0016] Step 3. Calculate the impedance value of the road based on the average speed of vehicles on the road and the current vehicle density of the road. The expression is:

[0017]

[0018] Where I is the impedance value, α and β are weight coefficients, α + β = 1, V is the average speed of vehicles on the road, V max is the maximum average speed of vehicles on the road, K is the vehicle density of the current road, K max is the maximum vehicle density of the current road;

[0019] Step 4. Analyze the rate of change of the impedance value with the average speed of vehicles on the road to determine the speed range for the autonomous vehicle to take over manually;

[0020] Step 5. When the average speed of vehicles on the road reaches the threshold of the speed range for manual takeover, the human-computer interaction module reminds the driver to take over.

[0021] As a preferred embodiment of the present invention, the car-following model of the autonomous driving vehicle is used to represent the motion mode of the autonomous driving vehicle, and the expression is:

[0022]

[0023] e n (t) = h n (t)-L-S0-t av V n (t);

[0024] Where V n (t+Δt) is the speed of the autonomous vehicle at time t+Δt, Δt is the time step, which is 0.01s, V n (t) is the speed of the autonomous vehicle at the current moment, k p is the spacing error control parameter, e n (t) is the error between the actual headway and the expected headway of the autonomous driving vehicle at the current moment, k d is the undetermined coefficient, for e n (t) the derivative with respect to time t, h n (t) is the distance between the autonomous vehicle and the vehicle in front, L is the length of the autonomous vehicle, S0 is the minimum safe distance, t av is the expected headway of the autonomous vehicle.

[0025] As a preferred embodiment of the present invention, the dynamic change coefficients in the following model of the manually driven vehicle and the following model of the autonomous driving vehicle are obtained by parameter fitting of the historical data stored in the database and the vehicle data acquired in real time, thereby obtaining the following model of the manually driven vehicle and the following model of the autonomous driving vehicle under different road environments. The dynamic change coefficients in the following model of the manually driven vehicle include the sensitivity coefficient λ, the undetermined coefficient m, the undetermined coefficients k and b, and the weight coefficient δ of the distance term; the dynamic change coefficients in the following model of the autonomous driving vehicle include the spacing error control parameter k p , undetermined coefficient k d , the expected headway time t of the autonomous vehicle av .

[0026] As a preferred embodiment of the present invention, the weight coefficients α and β in impedance value calculation are dynamically changing coefficients. In order to enable the calculated impedance value to be used normally in all current road environments, the weight coefficients α and β are determined as follows:

[0027] Step a. First, fix the values ​​of α and the ratio of autonomous vehicles to manually driven vehicles to obtain the minimum impedance value under the first road environment;

[0028] Step b. changing the ratio of autonomous vehicles to manually driven vehicles to obtain the minimum impedance value under the second road environment;

[0029] Step c. Repeat the operation of step b to obtain the minimum impedance value under all road environments at the current α value;

[0030] Step d. Based on the minimum impedance values ​​under all road conditions at the current α value, obtain the average value of the minimum impedance values ​​when α is fixed and the ratio of autonomous vehicles to manually driven vehicles changes;

[0031] Step e. Change the value of α and repeat steps a to d to obtain the new average value of the minimum impedance;

[0032] Step f. Taking the maximum average value of the minimum impedance value as the objective function to ensure the stability of the impedance value calculation, and finally obtaining the value of α.

[0033] As a preferred embodiment of the present invention, when step 4 analyzes the rate of change of the impedance value with the average speed of the vehicle on the road, if the rate of change is less than 0.5%, it is considered that the driving efficiency of the autonomous driving vehicle in the speed range of the rate of change <0.5% is relatively stable; conversely, if the rate of change is ≥0.5%, it is considered that the driving efficiency of the autonomous driving vehicle in the speed range of the rate of change ≥0.5% is low, and the driver is allowed to take over the driving in the speed range of the rate of change ≥0.5%.

[0034] The present invention also provides an automatic driving takeover system based on road impedance in an intelligent network environment, the system comprising a car-following model, an impedance value calculation model, a takeover interval determination module, and a human-computer interaction module;

[0035] The car-following model is used to determine the motion modes of manually driven vehicles and autonomous vehicles under different road environments. The car-following model includes a car-following model for manually driven vehicles and a car-following model for autonomous vehicles. The car-following model for manually driven vehicles is used to represent the motion modes of manually driven vehicles, and is expressed as:

[0036] a n (t) = λv n (t) m {v n-1 (t)-v n (t)+δ[x n-1 (t)-x n (t)-X]};

[0037] X=kv n (t)+b;

[0038] Where a n (t) is the acceleration of the following vehicle at the current moment, λ is the sensitivity coefficient, v n (t) is the speed of the following vehicle at the current moment, m is the coefficient to be determined, v n-1 (t) is the speed of the preceding vehicle at the current moment, δ is the weight coefficient of the distance term, and x n-1 (t) is the position of the preceding vehicle at the current moment, x n (t) is the position of the following vehicle at the current moment, X is the expected following distance of the following vehicle, k and b are unknown coefficients;

[0039] The car-following model of the autonomous driving vehicle is used to represent the movement mode of the autonomous driving vehicle;

[0040] The impedance value calculation model is used to quantify the interference of the road environment on the autonomous driving vehicle and convert the complex road environment into a calculable impedance value. The expression is:

[0041]

[0042] Where I is the impedance value, α and β are weight coefficients, α + β = 1, V is the average speed of vehicles on the road, V max is the maximum average speed of vehicles on the road, K is the vehicle density of the current road, K max is the maximum vehicle density of the current road;

[0043] The vehicle density K of the current road is determined based on the following model of manually driven vehicles and the following model of autonomous vehicles, and is expressed as:

[0044]

[0045] Where, t av is the expected headway of the autonomous vehicle, L is the length of the autonomous vehicle, S0 is the minimum safe distance, m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′+n=1, and V is the average speed of vehicles on the road;

[0046] The takeover interval determination module analyzes the rate of change of the impedance value with the average speed of vehicles on the road to obtain the operating efficiency of the autonomous driving vehicle at different average speeds, thereby determining the speed interval for manual takeover of the autonomous driving vehicle;

[0047] The human-computer interaction module is used to remind the driver to take over when the average speed of vehicles on the road reaches the threshold of the speed range for manual takeover, so as to ensure the reliability of the autonomous driving vehicle.

[0048] As a preferred embodiment of the present invention, the system further includes a computing platform, which is used to calculate the dynamic change coefficients in each model based on historical data and real-time acquired data, thereby determining the car-following model and the impedance value calculation model.

[0049] As a preferred embodiment of the present invention, the human-computer interaction module includes a tactile feedback module, a visual warning module and a sound prompt module. The tactile feedback module reminds the driver to take over manually by vibration; the visual warning module reminds the driver of the speed range for manual takeover and the average speed of vehicles on the current road through a display, so that the driver can understand the urgency of the takeover requirement; the sound prompt module reminds the driver to take over manually by voice broadcast.

[0050] The advantages and beneficial effects of the present invention are:

[0051] (1) The present invention takes into account that in an intelligent connected environment, data on the road environment can be obtained in real time, and proposes an automatic driving takeover system based on road impedance in an intelligent connected environment. The system can update the speed range for manual takeover in real time according to the road environment, and promptly remind the driver to perform the takeover operation at the takeover node, thereby improving the driver's user experience and the reliability of the automatic driving vehicle.

[0052] (2) The takeover system provided by the present invention can, with the help of a high-performance computing platform, efficiently process various data in the autonomous driving process in real time, thereby accurately judging whether the autonomous driving vehicle needs to be taken over under the current road environment, so as to avoid missing the best takeover opportunity, and at the same time facilitate the realization of higher-level autonomous driving functions.

[0053] (3) The takeover system provided by the present invention is based on human-computer interaction hardware, and realizes real-time feedback of driving status through a multimodal interaction channel. When confirming to take over, it can promptly remind the driver to perform the takeover operation through tactile feedback, visual warnings and sound prompts, thereby improving the fault tolerance of the automatic driving system and enabling the vehicle to maintain reliable operation under extreme system conditions.

[0054] (4) The takeover system provided by the present invention can define the speed range for manual takeover. Under different road conditions, the system can also adjust the speed range in real time. Existing takeover systems do not quantify the triggering conditions for manual takeover. The takeover system provided by the present invention specifies the nodes at which autonomous driving vehicles should be manually taken over under various road conditions, thereby promoting the development of autonomous driving technology.

[0055] (5) The car-following model for manually driven vehicles provided by the present invention is obtained by improving the GM car-following model. The original GM car-following model has only the speed as the stimulus item, which will cause the distance between the two vehicles to be too large or too small, which is not conducive to calculating the impedance value. Therefore, the present invention adjusts the stimulus items and expressions of the GM car-following model. The improved GM car-following model does not consider time delay, and the stimulus is determined by both speed and distance. The stimulus generated by the distance also considers the expected headway. This design is conducive to calculating the impedance value of the road, and the impedance value is used to represent the degree to which the driving of the autonomous driving vehicle is disturbed by the road environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is the logic block diagram of the autonomous driving takeover system based on road impedance in an intelligent connected environment;

[0057] Figure 2 is a graph of road impedance values ​​for all combinations of autonomous vehicles and manually driven vehicles;

[0058] Figure 3 It is the rate of change graph of road impedance value; DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the technical solutions and advantages of the present invention, the present application is described in detail below with reference to the accompanying drawings, but this is not intended to limit the scope of protection of the present invention.

[0060] Figure 1 This is a logic block diagram of an automatic driving takeover system based on road impedance in an intelligent network environment provided by the present invention, such as Figure 1 As shown, this embodiment provides an autonomous driving takeover system based on road impedance in an intelligent connected environment, comprising: a car-following model, an impedance value calculation model, a takeover interval determination module, and a human-computer interaction module;

[0061] The car-following model is used to determine the motion modes of manually driven vehicles and autonomous vehicles under different road environments; the car-following model includes a car-following model for manually driven vehicles and a car-following model for autonomous vehicles;

[0062] In this embodiment, the car-following model of the manually driven vehicle is used to represent the movement mode of the manually driven vehicle. The movement mode of the manually driven vehicle is:

[0063] a n (t) = λv n (t) m {v n-1 (t)-v n (t)+δ[x n-1 (t)-x n (t)-X]}

[0064] X=kv n (t)+b

[0065] Where a n (t) is the acceleration of the following vehicle at the current moment, λ is the sensitivity coefficient, v n (t) is the speed of the following vehicle at the current moment, m is the coefficient to be determined, v n-1 (t) is the speed of the preceding vehicle at the current moment, δ is the weight coefficient of the distance term, and x n-1 (t) is the position of the preceding vehicle at the current moment, x n (t) is the position of the following vehicle at the current moment, X is the expected following distance of the following vehicle, k and b are unknown coefficients, which are determined by parameter fitting the position information of the leading and trailing vehicles and the speed information of the following vehicle during stable following on the road;

[0066] The car-following model of the autonomous vehicle is used to represent the motion mode of the autonomous vehicle. The motion mode of the autonomous vehicle is:

[0067]

[0068] e n (t) = h n (t)-L-S0-t av V n (t)

[0069] Where V n (t+Δt) is the speed of the autonomous vehicle at time t+Δt, Δt is the time step, which is 0.01s, V n (t) is the speed of the autonomous vehicle at the current moment, k p is the spacing error control parameter, e n (t) is the error between the actual headway and the expected headway of the autonomous driving vehicle at the current moment, kd is the unknown coefficient, for e n (t) the derivative with respect to time t, h n (t) is the distance between the autonomous vehicle and the vehicle in front (following distance), L is the length of the autonomous vehicle, S0 is the minimum safe distance, t a is the expected headway of the autonomous vehicle;

[0070] The vehicle density of a road represents the number of vehicles per unit length. Since the following distances between manually driven and autonomous vehicles are different, and autonomous vehicles generally have shorter following distances, different road environments will result in different average vehicle speeds and vehicle densities. The driving of autonomous vehicles will be disturbed by the road environment, and this disturbance can be expressed as the road's impedance. Therefore, the present invention designs an impedance calculation model to quantify the interference of the road environment on autonomous vehicles. The impedance calculation model converts the complex road environment into a calculable impedance value, obtaining the impedance value of the autonomous vehicle at different speeds. The expression is:

[0071]

[0072] Where I is the impedance value, α and β are weight coefficients, α + β = 1, V is the average speed of vehicles on the road (autonomous driving vehicles and manually driven vehicles), V max is the maximum average speed of vehicles on the road, K is the vehicle density of the current road, K max is the maximum vehicle density of the current road;

[0073] In this embodiment, the vehicle density K of the current road is determined based on the car-following model of the manually driven vehicle and the car-following model of the autonomous driving vehicle, and is expressed as:

[0074]

[0075] Where, t av is the expected headway of the autonomous vehicle, L is the length of the autonomous vehicle, S0 is the minimum safe distance, m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′+n=1, and V is the average speed of vehicles on the road;

[0076] The takeover interval determination module analyzes the rate of change of the impedance value with the average speed of vehicles on the road to obtain the operating efficiency of the autonomous driving vehicle at different average speeds, thereby determining the speed interval for manual takeover of the autonomous driving vehicle, which is the appropriate takeover interval;

[0077] The human-computer interaction module is used to remind the driver to take over when the average speed of vehicles on the road reaches the threshold of the speed range for manual takeover, so as to ensure the reliability of the autonomous driving vehicle.

[0078] In this example, the dynamic change coefficients in the following model of the manually driven vehicle and the following model of the autonomous driving vehicle are obtained by parameter fitting of the historical data stored in the database and the vehicle data acquired in real time, thereby obtaining the following model of the manually driven vehicle and the following model of the autonomous driving vehicle on different roads. The dynamic change coefficients in the following model of the manually driven vehicle include the sensitivity coefficient λ, the undetermined coefficient m, the undetermined coefficients k and b, and the weight coefficient δ of the distance term; the dynamic change coefficients in the following model of the autonomous driving vehicle include the spacing error control parameter k p , undetermined coefficient k d , the expected headway time t of the autonomous vehicle av ; In addition, the weight coefficients α and β when calculating the impedance value are also dynamically changing coefficients.

[0079] Furthermore, in this example, the system also includes a high-performance computing platform, which is used to calculate the dynamic change coefficients in each model based on historical data and real-time acquired data, thereby determining the car-following model and the impedance value calculation model.

[0080] In this example, the human-computer interaction module includes a tactile feedback module, a visual warning module and a sound prompt module. The tactile feedback module reminds the driver to take over manually by vibration; the visual warning module reminds the driver of the speed range for manual takeover and the average speed of vehicles on the current road through a display, so that the driver can understand the urgency of the takeover requirement; the sound prompt module reminds the driver to take over manually by voice broadcast, ensuring that the autonomous driving vehicle interacts with the driver in real time.

[0081] In this example, the autonomous driving vehicle stores the trajectory data of manually driven vehicles and autonomous driving vehicles in different road environments. At the same time, the autonomous driving vehicle can also monitor and obtain the trajectory data of surrounding vehicles in real time in an intelligent network environment. Through these trajectory data, the parameters of the following model are fitted in real time, thereby obtaining the following model of manually driven vehicles and the following model of autonomous driving vehicles in different road environments.

[0082] Specifically, taking the NGSIM (Next Generation Simulation) project as an example, the specific road environment is vehicle data on Highway 101 in Los Angeles, California, USA. When calculating the values ​​of the dynamic change parameters of the following model of a manually driven vehicle, a high-performance computing platform is used to fit the data set, and the values ​​of the dynamic change parameters are determined to be: λ = 0.6, m = 0.1, δ = 0.1, k = 1.27, b = 7.78.

[0083] In this embodiment, the parameter values ​​of the car-following model of the autonomous driving vehicle can be determined by fitting on a high-performance computing platform, or the existing publicly reported model can be directly adopted; for example, k p =0.45, k d =0.25, L=5m, S0=2m, t av =0.6s.

[0084] In this embodiment, the parameter V in the impedance value calculation model is max The maximum speed on the actual road is determined. For example, if the speed limit on a highway is 120 km / h, the theoretical average speed of vehicles is between 0 and 120 km / h, and the maximum average speed is 120 km / h. Of course, the speed limit varies on different roads and can also be 26.27 m / s (the maximum vehicle speed in the NGSIM dataset is 26.27 m / s), meaning the maximum average speed on the road is 26.27 m / s.

[0085] K max It can be deduced from theory that in this embodiment, when the road vehicles are running stably, the acceleration of all vehicles is 0m / s 2 At this time, the expected headway (following distance) for manually driven vehicles is 1.27V+7.78, and the expected headway for autonomous vehicles is 0.6V+7. The vehicle density K of the road represents the number of vehicles per unit length, and is calculated as follows:

[0086]

[0087] Where m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′ + n = 1, and V represents the average speed of vehicles on the road.

[0088] According to the calculation formula of K, when V is 0m / s and n is 1 (when all vehicles on the road are self-driving cars), K is the maximum. The maximum value K max =142.86.

[0089] In this embodiment, the ratio of autonomous vehicles to manually driven vehicles on the road is uncertain. Under different ratios, the vehicle density on the road will change. In order to ensure that the impedance value calculation model can be used normally in all current road environments, the values ​​of α and β are obtained based on the robustness principle. The specific method is as follows:

[0090] Step a. First, fix the values ​​of α and the ratio of autonomous vehicles to manually driven vehicles to obtain the minimum impedance value under the first road environment;

[0091] Step b. changing the ratio of autonomous vehicles to manually driven vehicles to obtain the minimum impedance value under the second road environment;

[0092] Step c. Repeat the operation of step b to obtain the minimum impedance value under all road environments at the current α value;

[0093] Step d. Based on the minimum impedance values ​​under all road conditions at the current α value, obtain the average value of the minimum impedance values ​​when α is fixed and the ratio of autonomous vehicles to manually driven vehicles changes;

[0094] Step e. Change the value of α and repeat steps a to d to obtain the new average value of the minimum impedance;

[0095] Step f. Taking the maximum average value of the minimum impedance value as the objective function, the stability of the impedance value calculation module is ensured, and finally the value of α is obtained.

[0096] In this embodiment, the final value of α is 0.33. Since α+β=1, the value of β is 0.67. Among them, the larger the impedance value, the higher the efficiency of the autonomous driving.

[0097] In this embodiment, the takeover interval determination module calculates the rate of change of the impedance value of the autonomous driving vehicle at various speeds (the average speed of vehicles on the road) using an impedance function. If the rate of change is small (less than 0.5%), it is considered that the driving efficiency of the autonomous driving vehicle in this speed range is relatively stable; if the rate of change is large, it is considered that the driving efficiency of the autonomous driving vehicle in this speed range is low, and the driver is asked to take over the driving in this speed range. Figure 2 The impedance values ​​of autonomous vehicles and manually driven vehicles at all ratio combinations are shown. Figure 2 It can be seen that the operating efficiency of autonomous vehicles is poor when the speed (average speed of vehicles on the road) is too high or too low; Figure 2 The shaded area represents the impedance value when the ratio of autonomous vehicles to manually driven vehicles is different. The lower limit curve of the shaded area corresponds to the lowest operating efficiency. Figure 2 The impedance value change rate of the lower limit curve of the shaded part can be used to obtain the impedance value change rate diagram (such as Figure 3 shown), through Figure 3 It can be seen that within the speed range of 4.88 m / s to 15.81 m / s (the average speed of vehicles on the road), the impedance value has a low rate of change, with a variation of less than 0.5%; therefore, the manual driving ranges for autonomous vehicles are 0 m / s to 4.88 m / s and 15.81 m / s to 26.27 m / s.

[0098] In this example, the takeover interval determination module has a corresponding database under different road environments. The high-performance computing platform uses the database and real-time vehicle data to perform parameter fitting on the following models of manually driven vehicles and autonomous driving vehicles, and then uses the following model (different following models have different α and maximum density) to obtain the impedance function under different road environments, thereby determining the corresponding manual takeover speed interval; the human-computer interaction module promptly reminds the driver to take over according to the manual takeover speed interval to ensure the reliability of the autonomous driving vehicle.

[0099] This embodiment also provides a method for autonomous driving takeover based on road impedance in an intelligent connected environment, the method comprising the following steps:

[0100] Step 1. Construct a car-following model to determine the motion patterns of manually driven vehicles and autonomous vehicles under different road conditions. The car-following model includes a car-following model for manually driven vehicles and a car-following model for autonomous vehicles. The car-following model for manually driven vehicles is used to represent the motion patterns of manually driven vehicles, and is expressed as:

[0101] a n (t) = λv n (t) m {v n-1 (t)-v n (t)+δ[x n-1 (t)-x n (t)-X]};

[0102] X=kv n (t)+b;

[0103] Where a n (t) is the acceleration of the following vehicle at the current moment, λ is the sensitivity coefficient, v n (t) is the speed of the following vehicle at the current moment, m is the coefficient to be determined, v n-1 (t) is the speed of the preceding vehicle at the current moment, δ is the weight coefficient of the distance term, and x n-1 (t) is the position of the preceding vehicle at the current moment, x n(t) is the position of the following vehicle at the current moment, X is the expected following distance of the following vehicle, k and b are unknown coefficients, which are determined by parameter fitting the position information of the leading and trailing vehicles and the speed information of the following vehicle during stable following on the road;

[0104] The car-following model of the autonomous driving vehicle is used to represent the movement mode of the autonomous driving vehicle;

[0105] Step 2. Determine the average speed of vehicles on the road based on the current road environment, and determine the vehicle density of the road based on the car-following model for manually driven vehicles and the car-following model for autonomous vehicles.

[0106]

[0107] Among them, t av is the expected headway of the autonomous vehicle, L is the length of the autonomous vehicle, S0 is the minimum safe distance, m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′+n=1, and V is the average speed of vehicles on the road;

[0108] Step 3. Calculate the impedance value of the road based on the average speed of vehicles on the road and the current vehicle density of the road. The expression is:

[0109]

[0110] Where I is the impedance value, α and β are weight coefficients, α + β = 1, V is the average speed of vehicles on the road, V max is the maximum average speed of vehicles on the road, K is the vehicle density of the current road, K max is the maximum vehicle density on the current road, that is, the vehicle density when all vehicles on the road are self-driving vehicles;

[0111] Step 4. Analyze the rate of change of the impedance value with the average speed of vehicles on the road to determine the speed range for the autonomous vehicle to take over manually;

[0112] Step 5. When the average speed of vehicles on the road reaches the threshold of the speed range for manual takeover, the human-computer interaction module reminds the driver to take over.

[0113] The above description of the present invention using specific examples is intended only to facilitate understanding of the present invention and is not intended to limit the present invention. A person skilled in the art of the present invention may make several simple deductions, modifications, or substitutions based on the principles of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. An autonomous driving takeover method based on road impedance in an intelligent connected environment, characterized in that: The following steps are involved: Step 1. Construct a car-following model to determine the motion patterns of manually driven vehicles and autonomous vehicles under different road conditions. The car-following model includes both a manually driven vehicle and an autonomous vehicle. The car-following model of a manually driven vehicle is used to represent the motion of a manually driven vehicle, and is expressed as: a n (t)=λv n (t) m {v n-1 (t)-v n (t)+δ[x n-1 (t)-x n (t)-X]}; X=kv n (t)+b; Where a n (t) is the acceleration of the following vehicle at the current moment, λ is the sensitivity coefficient, v n (t) is the speed of the following vehicle at the current moment, m is the coefficient to be determined, v n-1 (t) is the speed of the preceding vehicle at the current moment, δ is the weight coefficient of the distance term, and x n-1 (t) is the position of the preceding vehicle at the current moment, x n (t) is the position of the following vehicle at the current moment, X is the expected following distance of the following vehicle, k and b are unknown coefficients; The car-following model of an autonomous vehicle is used to represent the movement of an autonomous vehicle; Step 2. Determine the average speed of vehicles on the road based on the current road environment, and determine the vehicle density of the road based on the car-following model for manually driven vehicles and the car-following model for autonomous vehicles. Where, t av is the expected headway of the autonomous vehicle, L is the length of the autonomous vehicle, S0 is the minimum safe distance, m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′+n=1, and V is the average speed of vehicles on the road; Step 3. Calculate the impedance value of the road based on the average speed of vehicles on the road and the current vehicle density of the road. The expression is: Where I is the impedance value, α and β are weight coefficients, α + β = 1, V is the average speed of vehicles on the road, V max is the maximum average speed of vehicles on the road, K is the vehicle density of the current road, K max is the maximum vehicle density of the current road; Step 4. Analyze the rate of change of the impedance value with the average speed of vehicles on the road to determine the speed range for the autonomous vehicle to take over manually; Step 5. When the average speed of vehicles on the road reaches the threshold of the speed range for manual takeover, the human-computer interaction module reminds the driver to take over.

2. The method for automatic driving takeover based on road impedance in an intelligent connected environment according to claim 1, characterized in that: The car-following model of the autonomous vehicle is used to represent the motion mode of the autonomous vehicle, and the expression is: e n (t)=h n (t)-L-S0-t av V n (t); Where V n (t+Δt) is the speed of the autonomous vehicle at time t+Δt, Δt is the time step, which is 0.01s, V n (t) is the speed of the autonomous vehicle at the current moment, k p is the spacing error control parameter, e n (t) is the error between the actual headway and the expected headway of the autonomous driving vehicle at the current moment, k d is the unknown coefficient, for e n (t) the derivative of time t, h n (t) is the distance between the autonomous vehicle and the vehicle in front, L is the length of the autonomous vehicle, S0 is the minimum safe distance, t av is the expected headway of the autonomous vehicle.

3. The method for automatic driving takeover based on road impedance in an intelligent connected environment according to claim 2, characterized in that: The dynamic change coefficients in the following model of the manually driven vehicle and the following model of the autonomous driving vehicle are obtained by parameter fitting of historical data stored in the database and vehicle data acquired in real time, thereby obtaining the following model of the manually driven vehicle and the following model of the autonomous driving vehicle under different road environments. The dynamic change coefficients in the following model of the manually driven vehicle include the sensitivity coefficient λ, the undetermined coefficient m, the undetermined coefficients k and b, and the weight coefficient δ of the distance term; the dynamic change coefficients in the following model of the autonomous driving vehicle include the spacing error control parameter k p , undetermined coefficient k d , the expected headway time t of the autonomous vehicle av .

4. The method for automatic driving takeover based on road impedance in an intelligent connected environment according to claim 2, characterized in that: The weight coefficients α and β in impedance calculation are dynamic coefficients. In order to ensure that the calculated impedance value can be used normally in all current road environments, the weight coefficients α and β are determined as follows: Step a. First, fix the values ​​of α and the ratio of autonomous vehicles to manually driven vehicles to obtain the minimum impedance value under the first road environment; Step b. changing the ratio of autonomous vehicles to manually driven vehicles to obtain the minimum impedance value under the second road environment; Step c. Repeat the operation of step b to obtain the minimum impedance value under all road environments at the current α value; Step d. Based on the minimum impedance values ​​under all road conditions at the current α value, obtain the average value of the minimum impedance values ​​when α is fixed and the ratio of autonomous vehicles to manually driven vehicles changes; Step e. Change the value of α and repeat steps a to d to obtain the new average value of the minimum impedance; Step f. Taking the maximum average value of the minimum impedance value as the objective function to ensure the stability of the impedance value calculation, and finally obtaining the value of α.

5. The method for automatic driving takeover based on road impedance in an intelligent connected environment according to claim 2, characterized in that: In step 4, when analyzing the rate of change of the impedance value with the average speed of vehicles on the road, if the rate of change is less than 0.5%, it is considered that the driving efficiency of the autonomous driving vehicle is relatively stable in the speed range where the rate of change is less than 0.5%. Conversely, if the rate of change is greater than or equal to 0.5%, it is considered that the driving efficiency of the autonomous driving vehicle is low in the speed range where the rate of change is greater than or equal to 0.5%, and the driver is allowed to take over driving in the speed range where the rate of change is greater than or equal to 0.5%.

6. The autonomous driving takeover system based on road impedance in an intelligent connected environment is characterized by: Including car-following model, impedance value calculation model, takeover interval determination module, and human-computer interaction module; The car-following model is used to determine the motion modes of manually driven vehicles and autonomous vehicles under different road environments. The car-following model includes a car-following model for manually driven vehicles and a car-following model for autonomous vehicles. The car-following model for manually driven vehicles is used to represent the motion modes of manually driven vehicles, and is expressed as: a n (t)=λv n (t) m {v n-1 (t)-v n (t)+δ[x n-1 (t)-x n (t)-X]}; X=kv n (t)+b; Where a n (t) is the acceleration of the following vehicle at the current moment, λ is the sensitivity coefficient, v n (t) is the speed of the following vehicle at the current moment, m is the coefficient to be determined, v n-1 (t) is the speed of the preceding vehicle at the current moment, δ is the weight coefficient of the distance term, and x n-1 (t) is the position of the preceding vehicle at the current moment, x n (t) is the position of the following vehicle at the current moment, X is the expected following distance of the following vehicle, k and b are unknown coefficients; The car-following model of the autonomous driving vehicle is used to represent the movement mode of the autonomous driving vehicle; The impedance value calculation model is used to quantify the interference of the road environment on the autonomous driving vehicle and convert the complex road environment into a calculable impedance value. The expression is: Where I is the impedance value, α and β are weight coefficients, α + β = 1, V is the average speed of vehicles on the road, V max is the maximum average speed of vehicles on the road, K is the vehicle density of the current road, K max is the maximum vehicle density of the current road; The vehicle density K of the current road is determined based on the following model of manually driven vehicles and the following model of autonomous vehicles, and is expressed as: Where, t av is the expected headway of the autonomous vehicle, L is the length of the autonomous vehicle, S0 is the minimum safe distance, m′ is the proportion of manually driven vehicles, n is the proportion of autonomous vehicles, m′+n=1, and V is the average speed of vehicles on the road; The takeover interval determination module analyzes the rate of change of the impedance value with the average speed of vehicles on the road to obtain the operating efficiency of the autonomous driving vehicle at different average speeds, thereby determining the speed interval for manual takeover of the autonomous driving vehicle; The human-computer interaction module is used to remind the driver to take over when the average speed of vehicles on the road reaches the threshold of the speed range for manual takeover.

7. The automatic driving takeover system based on road impedance in an intelligent connected environment according to claim 6, characterized in that: The automatic driving takeover system also includes a computing platform, which is used to calculate the dynamic change coefficients in each model based on historical data and real-time acquired data, thereby determining a following model and an impedance value calculation model.

8. The automatic driving takeover system based on road impedance in an intelligent connected environment according to claim 6, characterized in that: The human-computer interaction module includes a tactile feedback module, a visual warning module and a sound prompt module. The tactile feedback module reminds the driver to take over manually by vibration; the visual warning module reminds the driver of the speed range for manual takeover and the average speed of vehicles on the current road through a display; and the sound prompt module reminds the driver to take over manually by voice broadcast.

Images