Automatic driving decision-making method based on driving intensity

Abstract

The invention discloses an automatic driving decision-making method based on driving intensity, and the method comprises the steps: calculating the reachable space of a current vehicle based on the motion state and driving intensity of the vehicle, reducing the reachable space, and obtaining a drivable space; the method comprises the following steps: respectively constructing a vehicle operability evaluation model, an energy consumption level evaluation model and a following evaluation model, respectively solving the evaluation models by adopting a hierarchical binary optimization method, integrating evaluation results, selecting an optimal preview point from a drivable space, and determining the driving intensity of the optimal preview point. According to the method, the stability of multi-objective decision solving is improved, and the stability of the vehicle is ensured.

Description

Technical Field

[0001] The present invention relates to the field of automobile technology, and in particular to an automatic driving decision-making method based on driving intensity. Background Art

[0002] Since the beginning of the 21st century, self-driving cars are gradually moving from imagination to reality. Mature self-driving technology can greatly reduce the incidence of traffic accidents, ensure the safety of drivers and pedestrians, and optimize the operation of traffic flow. In the closed-loop system of people, vehicles and roads, the driver is the most unreliable factor. Therefore, the study of driver models is of great significance. Most existing driver models analyze decision-making and control in the vehicle coordinate system. The calculation results have no direct connection with the driving operation, cannot well describe the driver's steering and speed control operations, and are not highly compatible with the vehicle chassis, which has certain limitations.

[0003] Therefore, a new preview control method is needed that can better fit the vehicle chassis structure and human driving habits while ensuring handling stability. Summary of the invention

[0004] Aiming at the problem that the traditional driver model using lateral and longitudinal acceleration as decision variables is not consistent with the vehicle chassis structure and driving habits, an automatic driving decision method based on driving intensity is proposed, which simplifies the control model, improves comfort and is more in line with driving habits while meeting safety requirements.

[0005] To achieve the above object, the present invention provides an automatic driving decision method based on driving intensity, comprising:

[0006] Based on the vehicle's motion state and driving intensity, the accessible space of the current vehicle is calculated, and the accessible space is reduced to obtain a drivable space;

[0007] A vehicle operability evaluation model, an energy consumption level evaluation model and a followability evaluation model are respectively constructed, and the evaluation models are respectively solved using a hierarchical binary optimization method. Based on the comprehensive evaluation results, an optimal preview point is selected from the drivable space, and the driving intensity of the optimal preview point is determined.

[0008] Preferably, calculating the accessible space of the current vehicle includes:

[0009] Calculating the driving intensity of the vehicle, wherein the driving intensity is represented by a spatial change rate of the heading angle along the way and a change rate of the vehicle speed over time, wherein the spatial change rate of the heading angle along the way is represented by the vehicle curvature, and the change rate of the vehicle speed over time is represented by the tangential acceleration;

[0010] Analyze the curvature range of the current vehicle based on the steering wheel angle range, wheelbase and traditional ratio of the steering wheel to the steering wheel, and analyze the tangential acceleration change range based on the current vehicle power specifications;

[0011] The situation of vehicle side slip is excluded, the curvature range is reduced, the driving intensity variation range is evenly discretized, and each driving intensity is combined as the future motion state of the vehicle to form the reachable space.

[0012] Preferably, the vehicle curvature and the tangential acceleration are calculated as follows:

[0013]

[0014] Where θ is the heading angle, is the yaw angle, L is the vehicle wheelbase, R is the vehicle turning radius, k is the vehicle curvature, v is the vehicle speed, t is time, and a is the tangential acceleration;

[0015] The method for calculating the curvature range is:

[0016]

[0017] Where θ is the maximum steering wheel angle, n is the transmission ratio between the steering wheel and the steering wheel, k is max is the maximum curvature of the vehicle.

[0018] Preferably, the method for calculating the reachable space is:

[0019]

[0020] Where a is the magnitude of the tangential acceleration, h is the heading angle, x and y are the horizontal and vertical coordinates respectively, v is the speed of the vehicle, the subscripts i and i-1 represent any continuous current moment and previous moment, and Δt represents the time interval between the current moment and the previous moment.

[0021] Preferably, obtaining the drivable space includes:

[0022] In combination with the legality of the current vehicle operation, the solid line adjacent lanes of the opposite lane and the own lane in the reachable space are removed, and the space outside the stop line is removed when the traffic light is red;

[0023] Considering the driving safety of the current vehicle, the intersection position between the bounding box corresponding to the preview position of the current vehicle and the bounding boxes of other traffic participants is removed.

[0024] Preferably, the vehicle operability evaluation model is constructed based on the vehicle's speed regulation operability and steering operability, specifically:

[0025] J h =wa *J a +w k *J k

[0026] In the formula, J a is the evaluation index of speed regulation in operability, J k is the evaluation index of steering in operability, J h is the overall operability evaluation index, w a is the weight of the speed regulation evaluation index, w k is the weight of the steering evaluation index.

[0027] Preferably, the energy consumption level evaluation model is constructed based on the ergonomics and operability of the vehicle, specifically:

[0028] J v =w e *J e +w h *J h

[0029] In the formula, J e is the ergonomic evaluation index, J h is the overall operability evaluation index, J v is the overall energy consumption level evaluation index, w e is the weight of the ergonomic evaluation index, w h is the weight of the manipulability evaluation index.

[0030] Preferably, the following performance evaluation model is constructed based on the longitudinal following performance and the lateral following performance of the vehicle, specifically:

[0031] J s =w c *J c +w l *J l

[0032] In the formula, J c is the lateral followability evaluation index, J l is the longitudinal followability evaluation index, J s is the overall path following evaluation index, w c is the weight of the lateral followability evaluation index, w l is the weight of the longitudinal follow-up evaluation index.

[0033] Preferably, the method for synthesizing the evaluation results is:

[0034] J=w s *J s +w v *J v

[0035] In the formula, J s is the path following evaluation index, J v is the energy consumption level evaluation index, J is the overall evaluation index, and w s is the weight of the path following evaluation index, w v is the weight of the energy consumption level evaluation index.

[0036] Preferably, selecting the optimal preview point includes:

[0037] An adaptive grid optimization algorithm is used to solve the point with the minimum cost, that is, the optimal preview point, and determine the optimal preview driving intensity.

[0038] Compared with the prior art, the present invention has the following advantages and technical effects:

[0039] (1) The present invention uses driving intensity as a decision variable to decouple the speed regulation and steering of the vehicle, so that the preview process is more in line with human driving habits and more in line with the vehicle chassis structure, thus simplifying the vehicle motion prediction model;

[0040] (2) The present invention uses driving intensity as a decision variable, which is a more intuitive and direct control variable, can reduce the vibration and impact of the vehicle during turning and acceleration and deceleration, and improve the stability and comfort of driving;

[0041] (3) The present invention adopts an adaptive grid optimization algorithm for solving the problem, which improves the stability of the multi-objective decision-making solution and ensures the stability of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0043] Figure 1 is an Ackerman steering diagram of a vehicle in an embodiment of the present invention;

[0044] Figure 2 is a flow chart of a preview decision method in an embodiment of the present invention;

[0045] Figure 3 is a schematic diagram of an accessible space in an embodiment of the present invention;

[0046] Figure 4 Schematic diagram of a preview path in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0048] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0049] The present invention proposes an automatic driving decision-making method based on driving intensity, such as Figure 2 ,include:

[0050] Based on the vehicle's motion state and driving intensity, the accessible space of the current vehicle is calculated, and considering the legality and safety of the vehicle and traffic participants, the accessible space is reduced to obtain a drivable space;

[0051] A vehicle operability evaluation model, an energy consumption level evaluation model and a followability evaluation model are respectively constructed, and the evaluation models are respectively solved using a hierarchical binary optimization method. Based on the comprehensive evaluation results, an optimal preview point is selected from the drivable space, and the driving intensity of the optimal preview point is determined.

[0052] The present invention improves the stability of multi-objective decision-making solutions and ensures the stability of the vehicle.

[0053] Furthermore, the accessible space of the current vehicle is calculated, including:

[0054] Calculating the driving intensity of the vehicle, the driving intensity is represented by the spatial change rate of the heading angle along the way and the change rate of the vehicle speed over time, the spatial change rate of the heading angle along the way is represented by the vehicle curvature, and the change rate of the vehicle speed over time is represented by the tangential acceleration;

[0055] Analyze the curvature range of the current vehicle based on the steering wheel angle range, wheelbase and traditional ratio of the steering wheel to the steering wheel, and analyze the tangential acceleration change range based on the current vehicle power specifications;

[0056] The situation of vehicle side slip is excluded, the curvature range is reduced, the driving intensity variation range is evenly discretized, and each driving intensity is combined as the future motion state of the vehicle to form the reachable space.

[0057] Specifically, the calculation method of vehicle curvature and tangential acceleration is:

[0058]

[0059] Where θ is the heading angle, is the yaw angle, L is the vehicle wheelbase, R is the vehicle turning radius, k is the vehicle curvature, v is the vehicle speed, t is time, and a is the tangential acceleration;

[0060] The method for calculating the curvature range is:

[0061]

[0062] Where θ is the maximum steering wheel angle, n is the transmission ratio between the steering wheel and the steering wheel, k is max is the maximum curvature of the vehicle.

[0063] Furthermore, the method for calculating the reachable space is:

[0064]

[0065] Where a is the magnitude of the tangential acceleration, h is the heading angle, x and y are the horizontal and vertical coordinates respectively, v is the speed of the vehicle, the subscripts i and i-1 represent any continuous current moment and previous moment, and Δt represents the time interval between the current moment and the previous moment.

[0066] Furthermore, obtaining the drivable space includes:

[0067] In combination with the legality of the vehicle's operation, the oncoming lane and the solid lane adjacent to the own lane in the accessible space are removed, and the space outside the stop line is removed when the traffic light is red;

[0068] Considering the driving safety of the vehicle, the intersection position between the bounding box corresponding to the preview position of the current vehicle and the bounding boxes of other traffic participants is removed.

[0069] Specifically, for legality, it mainly includes the oncoming lane, solid lane change and stop line of traffic lights. According to the environmental road information, the oncoming lane and the solid lane adjacent to the own lane in the accessible space are removed, and the space outside the stop line is removed when the light is red; for safety, the collision with other traffic participants is mainly considered, and the polygon intersection detection method is used for judgment. When the bounding box corresponding to the preview position of the own vehicle intersects with the bounding box of other traffic participants, the position is considered unsafe and removed from the drivable space.

[0070] Furthermore, the vehicle operability evaluation model is constructed based on the vehicle's speed regulation operability and steering operability, specifically:

[0071] J h =w a *J a +w k *J k

[0072] In the formula, J ais the evaluation index of speed regulation in operability, J k is the evaluation index of steering in operability, w i is the weight of each evaluation index, J h is the overall operability evaluation index, w a is the weight of the speed regulation evaluation index, w k is the weight of the steering evaluation index.

[0073] Specifically, each evaluation index is normalized using the sigmoid function:

[0074]

[0075] In the formula, x is the normalized target, J is the normalized result, A and B are constants determined by the maximum and minimum values ​​of the normalized target respectively;

[0076] The vehicle's operability is evaluated based on the vehicle's acceleration change and curvature change, where A and B are calculated as follows:

[0077]

[0078]

[0079] In the formula, a max is the maximum tangential acceleration that the vehicle can currently achieve, a min is the minimum tangential acceleration, k max is the maximum curvature of the current vehicle without side sliding, k min is the minimum curvature.

[0080] Furthermore, the energy consumption level evaluation model is constructed based on the ergonomics and operability of the vehicle, specifically:

[0081] J v =w e *J e +w h *J h

[0082] In the formula, J e is the ergonomic evaluation index, J h is the overall operability evaluation index, w i is the weight of each evaluation index, J v is the overall energy consumption level evaluation index, w e is the weight of the ergonomic evaluation index, w h is the weight of the manipulability evaluation index.

[0083] Specifically, the ergonomics of the vehicle is evaluated based on the speed of the vehicle, wherein A and B are calculated as follows:

[0084]

[0085] In the formula, v max To meet the current maximum speed of the vehicle under traffic regulations, v min The vehicle can currently reach the minimum speed to comply with traffic regulations.

[0086] Furthermore, the following performance evaluation model is constructed based on the longitudinal following performance and the lateral following performance of the vehicle, specifically:

[0087] J s =w c *J c +w l *J l

[0088] In the formula, J c is the lateral followability evaluation index, J l is the longitudinal followability evaluation index, J s is the overall path following evaluation index, w c is the weight of the lateral followability evaluation index, w l is the weight of the longitudinal follow-up evaluation index.

[0089] Specifically, the longitudinal following ability of the vehicle is evaluated based on the angle between the vehicle and the center line of the reference lane, where A and B are calculated as follows:

[0090]

[0091] In the formula, θ max is the maximum angle, θ min is the minimum angle;

[0092] The lateral following ability of the vehicle is evaluated based on the distance between the vehicle and the center line of the reference lane, where A and B are calculated as follows:

[0093]

[0094] Where, d max is the maximum lateral distance, d min is the minimum lateral distance;

[0095] Furthermore, the method of comprehensively evaluating the results is:

[0096] J=w s *J s +w v *J v

[0097] In the formula, J s is the path following evaluation index, J vis the energy consumption level evaluation index, J is the overall evaluation index, and w s is the weight of the path following evaluation index, w v is the weight of the energy consumption level evaluation index.

[0098] An adaptive grid optimization algorithm is used to solve the point with the minimum cost, that is, the optimal preview point, and determine the optimal preview driving intensity.

[0099] Since multi-value optimization solutions tend to diverge, and it is easier to find the optimal solution by optimizing two values ​​with opposite effects, a hierarchical binary optimization method is used for solution. Since using larger motion changes often improves efficacy but reduces operability, efficacy and operability are combined for optimization to improve solution stability. Finally, considering the optimization results of the optimization and path following, the optimal preview driving intensity is comprehensively decided.

[0100] In order to more clearly express the technical solution of the present invention, the following specific embodiments are provided to introduce the solution:

[0101] This embodiment provides an automatic driving decision method based on driving intensity. The input in this embodiment is the dynamic driving environment around the vehicle and the vehicle motion state, and the output is the target driving intensity. The specific process includes:

[0102] Step 1: Calculate the accessible space of the current vehicle based on the vehicle's motion state and driving intensity.

[0103] Driving intensity refers to steering intensity and speed change intensity, which are expressed by the spatial change rate of heading angle and the time history change rate of vehicle speed, respectively. Figure 1 , calculated as follows:

[0104]

[0105] Where θ is the heading angle, is the yaw angle, L is the vehicle wheelbase, R is the vehicle turning radius, k is the vehicle curvature, v is the vehicle speed, t is time, and a is the tangential acceleration.

[0106] The vehicle curvature is used to represent the rate of change of the heading angle space along the way, which reflects the steering intensity of the steering wheel rotation;

[0107] Tangential acceleration is used to represent the rate of change of vehicle speed over time, which reflects the speed change intensity controlled by the accelerator and brake pedals.

[0108] First, the curvature range of the vehicle is analyzed based on the steering wheel angle range, wheelbase and traditional ratio of the steering wheel to the steering wheel, and the tangential acceleration variation range is analyzed based on the vehicle's power specifications.

[0109] The formula for calculating the curvature range based on the vehicle structure is as follows:

[0110]

[0111] Where θ is the maximum steering angle of the steering wheel, n is the transmission ratio between the steering wheel and the steering wheel, L is the vehicle wheelbase, k max is the maximum curvature of the vehicle.

[0112] Secondly, in order to eliminate the situation of vehicle side slip, the curvature range needs to be reduced. The final curvature takes the minimum value of the two as the range boundary. The calculation formula is as follows:

[0113]

[0114] In the formula, is the road adhesion coefficient, g is the acceleration of gravity, v is the vehicle speed, and 0.9 is the safety factor.

[0115] Finally, the above driving intensity variation range is uniformly discretized, and each driving intensity is combined as the future motion state of the vehicle to form a reachable space, such as Figure 3 The calculation formula is as follows:

[0116]

[0117] Where a is the magnitude of the tangential acceleration, k is the curvature of the vehicle, h is the heading angle, x and y are the horizontal and vertical coordinates respectively, v is the speed of the vehicle, the subscripts i and i-1 represent any continuous current moment and previous moment, and Δt represents the time interval between the current moment and the previous moment.

[0118] Step 2: Based on the legality and safety of the vehicle, consider traffic participants and reduce accessible space.

[0119] Regarding legality, it mainly includes the oncoming lane, the solid lane change and the stop line of the traffic light. According to the environmental road information, the oncoming lane and the solid lane adjacent to the own lane in the reachable space are removed, and the space outside the stop line is removed when the light is red; Regarding safety, the collision with other traffic participants is mainly considered, and the polygon intersection detection method is used for judgment. When the bounding box corresponding to the preview position of the own vehicle intersects with the bounding box of other traffic participants, the position is considered unsafe and removed from the reachable space.

[0120] Step 3: Establish an operability evaluation model based on the vehicle's speed regulation operability and steering operability.

[0121] Each evaluation index is normalized using the sigmoid function:

[0122]

[0123] In the formula, x is the normalized target, J is the normalized result, A and B are constants determined by the maximum and minimum values ​​of the normalized target respectively;

[0124] The vehicle's operability is evaluated based on the vehicle's acceleration change and curvature change, where A and B are calculated as follows:

[0125]

[0126] In the formula, a max is the maximum tangential acceleration that the vehicle can currently achieve, a min is the vehicle’s current minimum tangential acceleration, k max is the maximum curvature of the current vehicle without side sliding, k min It is the minimum curvature of the current vehicle without side sliding.

[0127] The operability evaluation index is as follows:

[0128] J h =w a *J a +w k *J k

[0129] In the formula, J a is the evaluation index of speed regulation in operability, J k is the evaluation index of steering in operability, J h is the overall operability evaluation index, w a is the weight of the speed regulation evaluation index, w k is the weight of the steering evaluation index.

[0130] Step 4: Establish an energy consumption level evaluation model based on the vehicle's ergonomics and operability.

[0131] The ergonomics of a vehicle is evaluated based on its speed, which reflects the length of time it takes for a vehicle to travel a distance. A and B are calculated as follows:

[0132]

[0133] In the formula, v max To meet the current maximum speed of the vehicle under traffic regulations, v min The vehicle can currently reach the minimum speed to comply with traffic regulations;

[0134] The energy consumption level evaluation index is as follows:

[0135] J v =w e *J e +w h *Jh

[0136] In the formula, J e is the ergonomic evaluation index, J h is the overall operability evaluation index, J v is the overall energy consumption level evaluation index, w e is the weight of the ergonomic evaluation index, w h is the weight of the manipulability evaluation index.

[0137] Step 5: Establish a followability evaluation model based on the longitudinal followability and lateral followability of the vehicle.

[0138] The longitudinal following ability of the vehicle is evaluated based on the angle between the vehicle and the center line of the reference lane, where A and B are calculated as follows:

[0139]

[0140] In the formula, θ max is the maximum angle, θ min is the minimum angle;

[0141] The lateral following ability of the vehicle is evaluated based on the distance between the vehicle and the center line of the reference lane, where A and B are calculated as follows:

[0142]

[0143] Where, d max is the maximum lateral distance, d min is the minimum lateral distance;

[0144] The path following evaluation index is as follows:

[0145] J s =w c *J c +w l *J l

[0146] In the formula, J s is the path following evaluation index, J v is the energy consumption level evaluation index, J is the overall evaluation index, and w s is the weight of the path following evaluation index, w v is the weight of the energy consumption level evaluation index.

[0147] Step 6: Based on the evaluation model, replace the point with the smallest cost in the achievable space as the driving target.

[0148] Since multi-value optimization solutions tend to diverge, and it is easier to find the optimal solution by optimizing two values ​​with opposite effects, a hierarchical binary optimization method is used for solution. Since using larger motion changes often improves efficacy but reduces operability, efficacy and operability are combined for optimization to improve solution stability. Finally, considering the optimization results of the optimization and path following, the optimal preview driving intensity is comprehensively decided.

[0149] The total cost of each point is calculated by combining various evaluation indicators as follows:

[0150] J=w s *J s +w v *J v

[0151] In the formula, J s is the path following evaluation index, J v is the energy consumption level evaluation index, J is the overall evaluation index, and w s is the weight of the path following evaluation index, w v is the weight of the energy consumption level evaluation index;

[0152] Adaptive grid optimization algorithm is used to solve the point with the minimum cost as the preview point, and the optimal preview driving intensity is determined, such as Figure 4 .

[0153] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An automatic driving decision-making method based on driving intensity, characterized in that: include: Based on the vehicle's motion state and driving intensity, the accessible space of the current vehicle is calculated, and the accessible space is reduced to obtain a drivable space; A vehicle operability evaluation model, an energy consumption level evaluation model and a followability evaluation model are respectively constructed, and the evaluation models are respectively solved using a hierarchical binary optimization method. Based on the comprehensive evaluation results, an optimal preview point is selected from the drivable space, and the driving intensity of the optimal preview point is determined.

2. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: Calculating the accessible space of the current vehicle, including: Calculating the driving intensity of the vehicle, wherein the driving intensity is represented by a spatial change rate of the heading angle along the way and a change rate of the vehicle speed over time, wherein the spatial change rate of the heading angle along the way is represented by the vehicle curvature, and the change rate of the vehicle speed over time is represented by the tangential acceleration; Analyze the curvature range of the current vehicle based on the steering wheel angle range, wheelbase and traditional ratio of the steering wheel to the steering wheel, and analyze the tangential acceleration change range based on the current vehicle power specifications; The situation of vehicle side slip is excluded, the curvature range is reduced, the driving intensity variation range is evenly discretized, and each driving intensity is combined as the future motion state of the vehicle to form the reachable space.

3. The automatic driving decision-making method based on driving intensity according to claim 2, characterized in that: The vehicle curvature and the tangential acceleration are calculated as follows: Where θ is the heading angle, is the yaw angle, L is the vehicle wheelbase, R is the vehicle turning radius, k is the vehicle curvature, v is the vehicle speed, t is time, and a is the tangential acceleration; The method for calculating the curvature range is: Where θ is the maximum steering wheel angle, n is the transmission ratio between the steering wheel and the steering wheel, k is max is the maximum curvature of the vehicle.

4. The automatic driving decision-making method based on driving intensity according to claim 3, characterized in that: The method for calculating the reachable space is: Where a is the magnitude of the tangential acceleration, h is the heading angle, x and y are the horizontal and vertical coordinates respectively, v is the speed of the vehicle, the subscripts i and i-1 represent any continuous current moment and previous moment, and Δt represents the time interval between the current moment and the previous moment.

5. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: The step of obtaining the drivable space includes: In combination with the legality of the current vehicle operation, the solid line adjacent lanes of the opposite lane and the own lane in the reachable space are removed, and the space outside the stop line is removed when the traffic light is red; Considering the driving safety of the current vehicle, the intersection position between the bounding box corresponding to the preview position of the current vehicle and the bounding boxes of other traffic participants is removed.

6. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: The vehicle operability evaluation model is constructed based on the vehicle's speed regulation operability and steering operability, specifically: J h =w a *J a +w k *J k In the formula, J a is the evaluation index of speed regulation in operability, J k is the evaluation index of steering in operability, J h is the overall operability evaluation index, w a is the weight of the speed regulation evaluation index, w k is the weight of the steering evaluation index.

7. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: The energy consumption level evaluation model is constructed based on the ergonomics and operability of the vehicle, specifically: J v =w e *J e +w h *J h In the formula, J e is the ergonomic evaluation index, J h is the overall operability evaluation index, J v is the overall energy consumption level evaluation index, w e is the weight of the ergonomic evaluation index, w h is the weight of the manipulability evaluation index.

8. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: The following performance evaluation model is constructed based on the longitudinal following performance and the lateral following performance of the vehicle, specifically: J s =w c *J c +w l *J l In the formula, J c is the lateral followability evaluation index, J l is the longitudinal followability evaluation index, J s is the overall path following evaluation index, w c is the weight of the lateral followability evaluation index, w l is the weight of the longitudinal follow-up evaluation index.

9. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: The method for synthesizing the evaluation results is: J=w s *J s +w v *J v In the formula, J s is the path following evaluation index, J v is the energy consumption level evaluation index, J is the overall evaluation index, and w s is the weight of the path following evaluation index, w v is the weight of the energy consumption level evaluation index.

10. The automatic driving decision-making method based on driving intensity according to claim 1, characterized in that: Selecting the optimal preview point includes: An adaptive grid optimization algorithm is used to solve the point with the minimum cost, that is, the optimal preview point, and determine the optimal preview driving intensity.

Images