Second-order Dynamic Authority Transfer Method Considering Drivers in the Case of Large-curvature Bend

Abstract

The present invention discloses a design method for a second-order dynamic human-machine authority transfer strategy considering drivers in the case of large-curvature bends, which relates to the field of intelligent driving technology and solves the problems that in the process of a driver who highly recognizes their own driving ability actively taking over the driving authority of a vehicle under the driving conditions of a large-curvature road, the traditional authority transfer strategy has a non-smooth transfer process and cannot be dynamically adjusted in real time according to the actual driving situation. The key points of its technical solution are to innovatively design a second-order dynamic authority transfer strategy by analogy with a second-order system, using driver steering information, vehicle speed, and road curvature parameters to achieve a real-time dynamically adjustable authority transfer process, making the authority transfer process match the actual needs. This authority transfer scheme takes into account both innovation and practicality. At the same time, this scheme uses the steering wheel angle and the steering wheel angular velocity to characterize the driver's steering information, greatly reducing the economic cost of the authority transfer strategy in vehicle applications and having broad commercial application prospects.

Description

Technical Field

[0001] This solution belongs to the technical field of intelligent driving, and relates to a method for second-order dynamic authority transfer considering drivers in the case of large-curvature curves. Background Art

[0002] Due to the complexity and dynamics of the traffic scene itself, there are potential legal and ethical issues in intelligent vehicle driving technology. Fully autonomous driving at levels L4-L5 is difficult to complete commercial promotion and application under current conditions. Currently, the mainstream technologies in the industry focus on assisted driving at levels L2-L3, and the vehicle needs to transfer the driving authority according to specific scenarios and requirements.

[0003] Existing research shows that for drivers with a high driving level, some of these drivers expect to take control of the vehicle's driving dominance in complex sections to maintain their recognition of their own driving dignity. These drivers aim to achieve the construction of self-expression through driving in complex sections. In this case, if the assisted driving system cannot smoothly transfer the vehicle control to the driver in a timely manner, it may trigger conflicts between the driver and the assisted driving system, thus affecting road traffic safety and going against the original intention of the design of the assisted driving system.

[0004] For the current mainstream L2-L3 level assisted driving systems, the authority transfer they adopt still focuses on simple stepwise or progressive transfer processes. This kind of authority transfer method can quickly and effectively complete the authority transfer, but there are problems such as insufficient smoothness in the transfer process, inability to accurately connect with the driver's steering condition, and inability to meet the needs of the driver's differential authority transfer process. The above problems may trigger certain human-machine conflicts during the authority transfer process, thus generating unstable steering operation behaviors and affecting the normal driving of the vehicle. Therefore, an authority transfer strategy that can combine the driver's steering state and the curved road conditions and achieve dynamic adjustment of authority transfer at a low economic cost is needed. Summary of the Invention

[0005] This solution proposes a method for second-order dynamic authority transfer considering drivers in the case of large-curvature curves. By using the real-time data of the driver's steering behavior combined with the vehicle speed and road curvature data, it realizes an overall smooth and dynamically time-varying vehicle driving authority transfer process from the assisted driving system to the driver, meeting the needs of drivers with high driving ability to dominate the vehicle through large-curvature curve paths. On the basis of low economic cost, it reduces the lateral deviation of the vehicle during the authority transfer process and improves the smoothness of vehicle driving.

[0006] A method for second-order dynamic authority transfer considering drivers in the case of large-curvature curves, the method includes:

[0007] When the vehicle enters a curve from a straight road and a permission transfer requirement from the assistance system to the driver is obtained, first, the driving permission is transferred to the driver through a second-order dynamic permission transfer strategy until the permission transfer amount reaches a set ratio, and then it turns into a linearized permission transfer until all the permissions are transferred to the driver;

[0008] The described second-order dynamic permission transfer strategy analogously describes the permission transfer process through a mass-spring-damper system. During the permission transfer process, the assistance system is regarded as an equivalent spring, which continuously releases permissions as a permission storage element, and the driver is regarded as an equivalent damper, which gradually receives permissions as a permission consumption element until the permission transfer amount reaches a set ratio, thereby constructing a basic model of the second-order dynamic permission transfer process;

[0009] The road curvature and vehicle speed are introduced into the above basic model;

[0010] The driver's steering behavior is used to construct a direct influence model of the second-order dynamic permission transfer process;

[0011] The driver's manipulation behavior is used to construct an indirect influence model of the second-order dynamic permission transfer process.

[0012] In the method of considering the driver's second-order dynamic permission transfer in the case of the above large-curvature curve, according to the driver's steering behavior, the above direct influence model is constructed by combining the driver's preview model.

[0013] In the method of considering the driver's second-order dynamic permission transfer in the case of the above large-curvature curve, the above indirect influence model is constructed based on the steering wheel angle and the steering wheel angular speed.

[0014] In the method of considering the driver's second-order dynamic permission transfer in the case of the above large-curvature curve, in the second-order dynamic permission transfer strategy, the assistance system is regarded as an equivalent spring, and the basic model in which the driver is regarded as an equivalent damper is expressed as:

[0015]

[0016] Where m represents the equivalent mass, which is the vehicle body mass here, b is the dynamic equivalent damping coefficient, k is the dynamic equivalent spring coefficient, λ is the current permission of the assistance system, represents the first derivative of λ; represents the second derivative of λ.

[0017] In the method of considering the driver's second-order dynamic permission transfer in the case of the above large-curvature curve, the road curvature and vehicle speed are introduced into the above basic model in the following way:

[0018]

[0019] is the natural frequency of the second-order dynamic authority transfer system, which is represented by the road curvature radius R and the vehicle longitudinal speed V x where a1 is an adjustable parameter. Since the equivalent mass m is known, the initial value of the equivalent spring coefficient k0 can be obtained. According to the definition of the critical damping system, b0 = 2mQ m the initial value of the equivalent damping coefficient b0 is obtained.

[0020] In the method for considering the driver's second-order dynamic authority transfer in the above-mentioned large-curvature bend situation, the characteristics of the direct influence model are as follows:

[0021]

[0022] where χ represents the degree of direct influence of the authority transfer on the driver's manipulation behavior, and R g is the transmission ratio of the vehicle steering system, and δ fd is the front wheel angle of the vehicle calculated according to the driver's preview model, which is used to characterize the steering behavior differences of different drivers. The driver's preview model is a reliable existing model and will not be elaborated here in detail.

[0023] In the method for considering the driver's second-order dynamic authority transfer in the above-mentioned large-curvature bend situation, the characteristics of the indirect influence model are as follows:

[0024]

[0025] Q s characterizes the degree of resistance of the driver's authority transfer, and Q sr characterizes the degree of adaptation of the driver's authority transfer, and θ sw is the steering wheel angle, is the square of the derivative of the steering wheel angle θ sw a 2,1 a 2,2 and a3 are adjustable parameters.

[0026] In the method for considering the driver's second-order dynamic authority transfer in the above-mentioned large-curvature bend situation, the linearized authority transfer strategy is as follows:

[0027]

[0028] represents the authority in the linear strategy process, and ~ is used to distinguish it from the authority in the second-order system.

[0029] In the method for considering the driver's second-order dynamic authority transfer in the above-mentioned large-curvature bend situation, the second-order dynamic authority transfer strategy is as follows:

[0030]

[0031] ρ r is the road curvature, and x dp is the driver's state variable, and T d and T p are the driver's preview time and reaction time respectively. a0 is a constant related to the driver's reaction time, and C lfn is the ratio of the preview distances at the near point and the far point of the vehicle, and K p is the expected steering proportional gain, and K c is the compensation steering ratio for the road surface curvature, and τ L is the differential time constant, and e ψL is the vehicle heading deviation, and e L is the vehicle lateral deviation.

[0032] In the above method of considering the driver's second-order dynamic authority transfer in the case of a large curvature bend, the comprehensive strategy of combining the second-order dynamic authority transfer and the linearized authority transfer of this method is as follows:

[0033]

[0034] ρ r is the road curvature, and x dp is the driver's state variable, and T d and T p are the driver's preview time and reaction time respectively. a0 is a constant related to the driver's reaction time, and C lfn = 0.4, which is the ratio of the preview distances at the near point and the far point of the vehicle, and K p is the expected steering proportional gain, and K c is the compensation steering ratio for the road surface curvature, and τ L is the differential time constant, and e ψL is the vehicle heading deviation, and e L is the vehicle lateral deviation.

[0035] The advantages of this solution are as follows:

[0036] 1) Since the authority transfer strategy is designed by combining the driver's steering behavior, vehicle speed, and road curvature information, the authority transfer strategy can be dynamically adjusted to match the driver's actual driving situation, which better meets the driver's actual needs, and the applicability of the authority transfer strategy has been greatly improved.

[0037] 2) The driver's steering behavior, vehicle speed, and road curvature information used in this solution are all easily obtainable data from the vehicle's built-in sensors, which greatly reduces the actual application cost of the authority transfer strategy and has broad application prospects and practicality in future assisted driving systems.

[0038] 3) This solution is obtained by analogical reasoning of a second-order system. The proposed authority transfer strategy has a simple structure and good versatility, and can be conveniently applied to different types of assisted driving systems. Description of the Drawings

[0039] Figure 1 It is a schematic diagram of the structure of the second-order dynamic authority transfer system of the present invention;

[0040] Figure 2 It is a schematic diagram of the application scenario of the second-order dynamic authority transfer system of the present invention;

[0041] Figure 3-1 It is a comparison chart of the lateral deviation of the vehicle under different authority transfer strategies in the simulation test;

[0042] Figure 3-2 It is a comparison chart of the heading deviation of the vehicle under different authority transfer strategies in the simulation test;

[0043] Figure 3-3 It is a comparison chart of the front wheel steering angle of the vehicle under different authority transfer strategies in the simulation test;

[0044] Figure 3-4 It is a comparison chart of the yaw rate of the vehicle under different authority transfer strategies in the simulation test;

[0045] Figure 4-1 It is a comparison chart of the lateral deviation of the vehicle under different authority transfer strategies in the driver-in-the-loop test;

[0046] Figure 4-2 It is a comparison chart of the heading deviation of the vehicle under different authority transfer strategies in the driver-in-the-loop test;

[0047] Figure 4-3 It is a comparison chart of the yaw rate of the vehicle under different authority transfer strategies in the driver-in-the-loop test. Detailed Embodiment

[0048] The technical solution of the present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

[0049] The present invention proposes a design method for a second-order dynamic authority transfer strategy considering the driver in the case of large-curvature curves, which is used to realize the real-time dynamic transfer of vehicle control authority from the vehicle to the driver in the case of large-curvature curve road conditions. In view of the need for high-level driving ability drivers to dominate the vehicle to pass through large-curvature curve paths, the present invention uses a second-order system to realize the real-time dynamic authority transfer process, and uses the sensors commonly configured on the vehicle to realize the real-time dynamic adjustment of the authority transfer process for the driver, vehicle, and road conditions at a relatively low economic cost, improving the ride comfort of the vehicle during the authority transfer process and the path tracking accuracy of the vehicle, and thus improving the driving experience of high-level drivers.

[0050] This solution adopts a second-order system as the main structure of the permission transfer strategy. In the scenarios considered in this solution, the auxiliary system retains all the permissions for vehicle driving in the initial state. After the driver initiates a permission transfer request, the vehicle driving permissions are transferred to the driver in real-time and dynamically. At the same time, to prevent overshoot, in cooperation with the linear transfer strategy, the transfer is carried out by the linear transfer strategy in the last certain proportion until the vehicle is completely controlled by the driver for driving. The way to initiate the permission transfer request and how the system detects or obtains this request are not within the scope of this solution and will not be described here. This solution can be used to solve the permission transfer problem after the transfer request is initiated in any way. Since the vehicle control permission transfer is unidirectional attenuation and continuously adjusted, the mass-spring-damper system can be used for analogical extension to realize the modeling description of the permission transfer process. In this system model, the vehicle body mass is regarded as the equivalent mass of the system, the starting point of the permission transfer is regarded as the peak value of the system energy, and the end point of the permission transfer is the equilibrium point of the system. Therefore, as Figure 1 shown, during the permission transfer process, the auxiliary system can be regarded as an equivalent spring, continuously releasing permissions as a permission storage element, while the driver is regarded as an equivalent damper, gradually receiving permissions as a permission consumption element. When all the permissions of the auxiliary system are consumed, the vehicle permissions are transferred to the driver, and the system returns to the equilibrium point accordingly.

[0051] The basic form corresponding to the second-order dynamic permission transfer strategy is as follows:

[0052]

[0053] Where m is the equivalent mass, b is the equivalent damping coefficient, k is the equivalent spring coefficient, and these two coefficients are dynamic values. λ is the permission of the current auxiliary system (value range: [0, 1]). During the entire permission transfer process, transferring the vehicle driving permissions to the driver is a process where λ continuously decreases from 1 to 0.

[0054] This solution takes into account the dynamic characteristics of the second-order system, makes the second-order dynamic permission transfer strategy change dynamically according to the driving situation, and establishes a relationship between the permission transfer process and the road curvature, vehicle speed, and driver steering state, as follows:

[0055] The natural frequency Q of the second-order dynamic permission transfer system is represented by the road curvature radius and the vehicle longitudinal speed m , where a1 is an adjustable parameter determined by the technical personnel according to the actual situation. For example, it can be determined as 2, and the specific value is not restricted here. R is the curvature radius. Since the vehicle body mass m is used as the vehicle body mass, the initial value k0 of the equivalent spring coefficient of this system can be obtained, and the initial value b0 of the equivalent damping coefficient is derived according to the definition of the critical damping system as follows:

[0056]

[0057] Meanwhile, in order to achieve dynamic real-time changes in authority transfer and enhance the understanding of the driver's steering characteristics during the authority transfer process, this solution introduces the driver's steering angle information as a characteristic into the design of the second-order dynamic authority transfer strategy, and incorporates steering behavior into the design process of the second-order dynamic authority transfer in both direct and indirect forms.

[0058] The direct adjustment part reflects the driver's authority transfer style:

[0059]

[0060] where χ represents the degree of direct influence of authority transfer on the driver's manipulation behavior. It is an empirical coefficient set in advance by technicians, R g is the transmission ratio of the vehicle steering system, δ fd is the front wheel steering angle of the vehicle, which is calculated according to the parameters of the two-point preview driver model and is used to characterize the steering behavior differences of different drivers.

[0061] The indirect adjustment further dynamically adjusts the authority transfer process by adjusting the equivalent spring coefficient and damping coefficient. These two coefficients are adjusted in real time according to the driver's participation degree in the operation. For the steering wheel angle θ sw size, Q s is used to characterize the driver's resistance degree to authority transfer. That is, under a certain road curvature, the larger the steering wheel angle, the smaller the driver's resistance degree to authority transfer. a 2,1 and a 2,2 are adjustable parameters, both determined by technicians according to specific situations. For example, they can be determined as a 2，1 = 5, a 2，2 = 10. There is no specific limitation here:

[0062]

[0063] The steering wheel angular velocity reflects the intensity of the driver's steering. When a skilled driver drives steadily on a large-curvature bend, the angular velocity of the steering wheel should tend to be gentle. Q sr is used to characterize the driver's adaptation degree to authority transfer. That is, under a certain road curvature, the larger the steering wheel angular velocity, the worse the driver's adaptation state to authority transfer. Among them, a3 is an adjustable parameter determined by technicians according to specific situations. For example, a3 = 5 can be determined:

[0064]

[0065] When the second-order dynamic permission transfer reaches the set ratio, it is completed, and then it turns into a linear permission transfer process, thus improving the smoothness of permission transfer. The set ratio is determined by technicians according to the actual situation. For example, it can be 95%, and there is no specific limit here.

[0066] Therefore, the second-order dynamic permission transfer strategy considering the driver proposed in this solution can be represented by the following state-space equation:

[0067]

[0068] In the formula, ρ r is the road curvature, which is equal to the reciprocal of the curvature radius R, x dp is the driver state variable, which is a process variable and is derived through the state-space equation (6) based on other parameters. T d and T p are the driver's preview time and reaction time, and their values can refer to some parameters in existing research. For example, Td = 0.14 and Tp = 1.2 can be taken. a0 is a constant related to the driver's reaction time. C lfn = 0.4 is the ratio of the preview distances of the vehicle at the near point and the far point. K p is the expected steering ratio gain. K c is the compensation steering ratio for road surface curvature. τ L is the differential time constant. e ψL is the vehicle heading deviation. e L is the vehicle lateral deviation. Some set values or parameters used in the formula are determined by technicians in this field according to the actual situation or existing research, and there is no specific limit here. For example, C lfn can be taken as 0.4, K p can be taken as 3.1, K c can be taken as 2.8, and so on.

[0069] The subsequent linear permission transfer strategy is designed as follows:

[0070]

[0071] represents the permission in the linear strategy process, and ~ is used to distinguish it from the permission in the second-order system.

[0072] Application scenario examples are as Figure 2As shown, the basic scenario built is a combination of a straight road and a U-turn road. During the straight driving stage, the control authority of the vehicle is dominated by the assisted driving system. After the vehicle enters the curve, based on the driver's dominant demand, the authority transfer is carried out, transferring the driving authority of the vehicle to the driver, and the driver completes the driving task. The curve is the starting point of the authority transfer. Through the second-order dynamic authority transfer strategy considering the driver designed by the present invention, the driving authority is transferred to the driver, realizing an authority transfer scheme for the real-time dynamic adjustment of the driving situation and driving scenario, and enhancing the smoothness of the authority transfer process and the driver's driving experience.

[0073] To prove the feasibility of this solution and its advantages over the existing technology transfer strategy, several different strategies are simulated and compared below.

[0074] Figures 3-1 to 3-4 It is a comparison chart of various parameters for the authority transfer achieved by the step transfer strategy, the asymptotic transfer strategy, and the transfer strategy of this solution under the Matlab / Simulink-Carsim joint simulation environment in the road condition with a large curvature of 0.01.

[0075] As can be seen from each figure, when entering a large curvature curve (curvature radius = 100m), the control authority is transferred from the automatic controller to the driver.

[0076] From Figure 3-1 it can be observed that compared with the step transfer, the second-order dynamic transfer has a smaller lateral deviation at the beginning of the transfer, indicating that starting from the moment of entering the curve, the second-order dynamic transfer strategy can use the robust path tracking controller to guide the vehicle to drive and gradually transfer the authority to the human driver, reducing the lateral deviation of the vehicle during the authority transfer process. Compared with the asymptotic transfer, the second-order dynamic transfer strategy can meet the dominant driving needs of the human driver faster and is closer to the actual needs of the driver. Therefore, the second-order dynamic transfer strategy can realize the guidance of the automatic controller to the driver during the authority transfer process, reduce the lateral deviation at the initial moment of the transfer, and at the same time quickly meet the dominant driving needs of the driver.

[0077] From Figure 3-2 it can be observed that compared with the step transfer, the second-order dynamic transfer strategy has a smaller heading deviation, indicating that the vehicle is closer to the desired yaw angle. Comparing the second-order dynamic transfer strategy with the asymptotic transfer, the change of the heading deviation is more stable, indicating that the vehicle drives more smoothly during the authority transfer process of the second-order dynamic transfer strategy. And Figure 3-3 shows that the peak value of the front wheel steering angle change of the vehicle under the second-order dynamic transfer strategy is less than that of the step transfer, and the smoothness of the steering angle change is better than that of the step transfer. This is consistent with Figure 3-4The trend of the vehicle's yaw rate shown is the same. The second-order dynamic authority transfer strategy makes the vehicle's yaw rate smoother, and there is no obvious yaw angle jitter after the authority transfer starts.

[0078] From the above simulation results, it can be seen that the proposed second-order dynamic authority transfer strategy is smoother during the authority transfer process compared to the other two strategies. This is because the second-order dynamic authority transfer strategy can make real-time dynamic adjustments according to the driver's steering behavior, while the other two authority transfer strategies can only transfer according to fixed strategies and there are authority mutations to varying degrees, which in turn affects the driving stability of the vehicle during the authority transfer process. Therefore, the second-order dynamic transfer strategy can keep the front wheel angle smooth with a small change amplitude, achieve a smoother change in the vehicle's yaw rate, and further ensure the stability of the vehicle during the authority transfer process.

[0079] Figures 4-1 to 4-3 The comparison results of vehicle driving by a test person under three authority transfer strategies are shown.

[0080] As Figures 4-1 to 4-3 shown, when entering a curve, there are small oscillations in both the step and progressive strategies, indicating that the driver has not fully adapted to the authority transfer process. The proposed second-order dynamic transfer strategy can achieve stable vehicle driving in the curve, and its lateral deviation, heading deviation, and yaw rate all remain at small values, indicating that the vehicle can maintain stable driving during the authority transfer under the second-order dynamic authority transfer strategy, and this transfer strategy is more acceptable to the driver.

[0081] Under the three different authority transfer strategies, the subjective evaluations of the participating test drivers on the driving process are shown in Table 1, and the analysis of variance (ANOVA) is shown in Table 2. The subjective evaluation is in the form of the driver's score, which is used to reflect the real feelings of the driver during the test. A 7-level Likert scale is used to statistically analyze the subjective feeling degree of the driver during the authority transfer process, where 0 represents the worst and 6 represents the best. Through the analysis of variance, it can be found that different authority transfer strategies have a significant impact on the subjective evaluation of the test subjects. Therefore, the designed second-order dynamic authority transfer strategy is effective. The subjective evaluations of the test personnel give significantly better evaluations for the second-order dynamic authority transfer, indicating that the participating drivers have obtained a better subjective driving experience through the second-order dynamic authority transfer strategy during the process of taking over vehicle driving in the curve, which is in line with the design intention of the second-order dynamic authority transfer strategy.

[0082] Table 1 Average value of subjective evaluations of test subjects

[0083]

[0084] Table 2 Analysis of variance (ANOVA) of subjective evaluations of test subjects

[0085]

[0086] **: p < 0.01

[0087] The specific embodiments described herein are merely illustrative of the spirit of the present solution. Those skilled in the art to which the present solution pertains can make various modifications or supplements to the described specific embodiments or use similar ways for substitution, but will not deviate from the spirit of the present solution or exceed the scope defined by the appended claims.

Claims

1. A second-order dynamic authority transfer method considering drivers in the case of a large-curvature bend, characterized in that The method includes: When the vehicle enters a curve from a straight road and a permission transfer requirement from the assistance system to the driver is obtained, first, the driving permission is transferred to the driver through a second-order dynamic permission transfer strategy until the permission transfer amount reaches a set ratio, and then it turns into a linearized permission transfer until all the permissions are transferred to the driver. The second-order dynamic permission transfer strategy describes the permission transfer process by analogy with a mass-spring-damper system. During the permission transfer process, the assistance system is regarded as an equivalent spring, which continuously releases permissions as a permission storage element, and the driver is regarded as an equivalent damper, which gradually receives permissions as a permission consumption element until the permission transfer amount reaches a set ratio, thereby constructing a basic model of the second-order dynamic permission transfer process. The road curvature and vehicle speed are introduced into the basic model. The driver's steering behavior is used to construct a direct influence model of the second-order dynamic permission transfer process. The driver's manipulation behavior is used to construct an indirect influence model of the second-order dynamic permission transfer process. In the second-order dynamic permission transfer strategy, the basic model in which the assistance system is regarded as an equivalent spring and the driver is regarded as an equivalent damper is expressed as: （1） wherein m represents the equivalent mass b is the equivalent damping coefficient k is the equivalent spring coefficient λ is the authority of the current auxiliary system represents λ the first derivative of represents λ the second derivative of The road curvature and vehicle speed are introduced into the basic model in the following way: （2） is the natural frequency of the second-order dynamic permission transfer system, and is represented by the road curvature radius and the vehicle longitudinal speed where a 1 is an adjustable parameter. Since the equivalent mass m is known, the initial value of the equivalent spring coefficient k 0 can be obtained. According to the definition of the critical damping system the initial value of the equivalent damping coefficient b 0 is obtained; The characteristics of the direct influence model are: （3） wherein represents the degree to which the authority transfer is directly affected by the driver's manipulation behavior, is the transmission ratio of the vehicle steering system, is the front wheel angle of the vehicle calculated according to the driver preview model, and is used to characterize the steering behavior differences of different drivers; The characteristics of the indirect influence model are: （4） Characterize the degree of conflict in driver authority transfer, Characterize the degree of adaptation in driver authority transfer, θ sw is the steering wheel angle, is the steering wheel angle θ sw the square of the derivative, a 2,1， a 2,2 and a 3 is an adjustable parameter.

2. The method for considering the second-order dynamic authority transfer of a driver in the case of a large-curvature bend according to claim 1, characterized in that The direct influence model is constructed based on the driver's steering behavior in combination with the driver preview model.

3. The method for considering the second-order dynamic authority transfer of the driver in the case of a large-curvature bend according to claim 2, characterized in that The indirect influence model is constructed based on the steering wheel angle and the steering wheel angular speed.

4. The method for considering the second-order dynamic authority transfer of the driver in the case of a large-curvature bend according to claim 3, characterized in that, The linearized permission transfer strategy is: （5） Represents the permissions in the linear policy process.

5. The method for considering the second-order dynamic authority transfer of the driver in the case of a large-curvature bend according to claim 4, characterized in that, The second-order dynamic permission transfer strategy is as follows: （6） is the road curvature, is the driver state quantity, T d and T p are the preview time and reaction time of the driver, a 0 is a constant related to the driver's reaction time, C lfn is the ratio of the preview distances of the vehicle at the near point and the far point, K p is the expected steering ratio gain, K c is the compensation steering ratio for the road surface curvature, is the differential time constant, is the vehicle heading deviation, is the vehicle lateral deviation.

6. The method for considering the second-order dynamic authority transfer of the driver in the case of a large-curvature bend according to claim 5, characterized in that The comprehensive strategy of this method combining the second-order dynamic permission transfer and the linearized permission transfer is as follows: is the road curvature, is the driver's state variable, T d and T p are the driver's preview time and reaction time, a 0 is a constant related to the driver's reaction time, C lfn is the ratio of the preview distances of the vehicle at the near point and the far point, K p is the expected steering ratio gain, K c is the compensation steering ratio for the road surface curvature, is the differential time constant, is the vehicle heading deviation, is the vehicle lateral deviation.

Images