A simulated driving evaluation method

Abstract

This invention provides a simulated driving evaluation method, comprising: Step 1, setting an information input area for preset control operations on the simulated driving information input terminal; Step 2, calibrating the correspondence between the preset control operations and real vehicle control operations; Step 3, acquiring a road image ahead, loading the simulated vehicle and initial driving parameters, receiving current location information, and establishing a simulated road environment; Step 4, acquiring and saving the input information of the preset control operations; Step 5, receiving a simulated driving termination command, and outputting an evaluation result based on the saved input information of the preset control operations. This invention realizes a simulated driving evaluation method in a simulated driving environment based on a real road environment, improving the realism of simulated driving.

Description

Technical Field

[0001] This invention relates to the field of driver assistance technology, and in particular to a method for evaluating simulated driving. Background Technology

[0002] With the continuous development of display technology, smart car cockpits are beginning to be equipped with instrument panel screens, central control screens, and passenger-side screens to meet the needs of drivers and passengers. The instrument panel screen primarily displays basic vehicle information and driving data, while the central control screen and passenger-side screen can display navigation information and provide some in-vehicle entertainment functions, such as video playback, music playback, and radio. However, current in-vehicle applications are not widespread, meaning that the passenger can only enjoy multimedia audio and video content, lacking interaction with the in-vehicle system.

[0003] Therefore, the existing technology needs further improvement. Summary of the Invention

[0004] This invention provides a simulated driving evaluation method, which aims to overcome the shortcomings of the prior art and realize a simulated driving evaluation method in a simulated driving environment based on real road environment, thereby improving the realism of simulated driving.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] This invention provides a simulated driving evaluation method, comprising:

[0007] Step 1: Set the preset control operation information input area on the simulated driving information input terminal;

[0008] Step 2: Calibrate the correspondence between the preset control operations and the actual vehicle control operations;

[0009] Step 3: Acquire an image of the road ahead, load the simulated vehicle and initial driving parameters, receive current location information, and establish a realistic road environment;

[0010] Step 4: Obtain and save the input information for the preset control operation;

[0011] Step 5: Receive the simulated driving termination command and output the evaluation result based on the saved input information of the preset control operation.

[0012] Specifically, step 3 includes:

[0013] Step 301: Identify road objects in the road image, including static road objects and dynamic road objects;

[0014] Step 302: Simplify the simulated vehicle and the road object according to preset rules;

[0015] Step 303: Load the simplified road object and its attributes, load the simulated vehicle and initial driving parameters, receive the current location information, and establish a realistic road environment.

[0016] Specifically, step 5 includes:

[0017] Step 501: Determine whether the simulated vehicle has a first preset behavior. If yes, the evaluation result is judged as unqualified. Otherwise, proceed to the next step. The first preset behavior is contact with a road object whose road object attribute is untouchable.

[0018] Step 502: Calculate the acceleration and change within the first preset range, and determine whether the change in acceleration is greater than the threshold. If it is, the evaluation result is determined to be unqualified; otherwise, proceed to the next step. The first preset range is the preset distance before and after the traffic light and the pedestrian crossing.

[0019] Step 503: Calculate fuel consumption based on the acceleration;

[0020] Step 504: Calculate the comprehensive evaluation value based on the acceleration change and fuel consumption.

[0021] Specifically, the first preset range is 20 to 100 meters.

[0022] Specifically, step 503 includes: when acceleration a≤0, fuel consumption F=1; when 0<acceleration a<1.2, fuel consumption F=4.17a+1; when acceleration a>1.2, fuel consumption F=7.

[0023] Specifically, the comprehensive evaluation value is calculated according to the following formula: P=k1 / Δa+k2*F, where P represents the comprehensive evaluation value, and k1 and k2 are empirical coefficients.

[0024] Specifically, determining whether the simulated vehicle exhibits a first preset behavior in step 501 includes:

[0025] Step A1: Obtain the origin and coordinates of each vertex of the simulated vehicle and the i-th road object in the world coordinate system;

[0026] Step A2: Calculate the distance Di between the origin of the simulated vehicle and the center of the i-th road object;

[0027] Step A3: Calculate the maximum collision detection radius R0(max) and minimum collision detection radius R0(min) of the simulated vehicle, and calculate the maximum collision detection radius Ri(max) and minimum collision detection radius Ri(min) of the i-th road object;

[0028] Step A4: Determine whether Di≤R0(max)+Ri(max) is true. If yes, proceed to the next step; otherwise, determine that no collision has occurred.

[0029] Step A5: Determine whether Di≤R0(min)+Ri(min) is true. If yes, it means that the two have collided; otherwise, proceed to the next step.

[0030] Step A6: Determine whether the simulated vehicle or the i-th road object has at least one vertex within the bounding box of the other. If so, determine that a collision has occurred; otherwise, determine that no collision has occurred.

[0031] Specifically, step A6, determining whether the simulated vehicle or the i-th road object has at least one vertex within the bounding box of the other side, includes:

[0032] Step A6-1: Take any vertex of the simulated vehicle or the i-th road object, and denote it as test vertex P(k);

[0033] Step A6-2: Represent the angle formed by the test vertex P(k) and the two adjacent vertices of the other side as the interior angle ∠Q(m)P(k)Q(m+1), where m=1,2,3…n-1 (n represents the number of vertices of the other side's border);

[0034] Step A6-3: Calculate the sum J of all the interior angles;

[0035] Step A6-4: Determine whether the sum of the angles J is equal to 360°. If so, determine that the test vertex P(k) is located within the border of the other side.

[0036] The beneficial effects of this invention are as follows: This invention sets preset control operations on the simulated driving information input terminal and calibrates the correspondence between the preset control operations and the real vehicle control operations. It also acquires images of the road ahead in real time to establish a simulated road environment. Then, based on the input information of the preset control operations, it outputs evaluation results, thereby realizing a simulated driving evaluation method in a simulated driving environment based on a real road environment, which improves the realism of simulated driving. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating the simulated driving evaluation method of the present invention;

[0038] Figure 2 This is a schematic diagram of the information input area of ​​the present invention. Detailed Implementation

[0039] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The drawings are for reference and illustration only and do not constitute a limitation on the scope of protection of the present invention.

[0040] Example 1

[0041] like Figure 1 As shown, this embodiment provides a simulated driving evaluation method, including:

[0042] Step 1: Set the preset control operation information input area on the simulated driving information input terminal.

[0043] In this embodiment, the simulated driving information input terminal can be a touch screen fixedly installed in front of the passenger seat of the car, or it can be a mobile terminal, such as the user's mobile phone.

[0044] The preset control operations include steering wheel control, accelerator control, gear control, handbrake light control, brake control, and turn signal control.

[0045] For example, such as Figure 2 As shown, the information input area 201 for steering wheel control can be set in the upper left corner, and the information input areas 202 for turn signal control can be set on both sides of the steering wheel. The information input areas 203 for brake pedal control and 204 for accelerator pedal control can be set below the information input area 201 for steering wheel control. The information input areas 205 for gear control and 206 for handbrake setting control can be set to the right of the information input area for steering wheel control. The upper right corner is set as the status information display area 207.

[0046] Step 2: Calibrate the correspondence between the preset control operations and the actual vehicle control operations.

[0047] For example, drawing clockwise or counterclockwise circles in the steering wheel control input area 201 can simulate the actual rotation of the steering wheel. However, this operation is often not sensitive enough; for example, rotating the wheel 10° on a touchscreen is very difficult to control. Therefore, it is necessary to calibrate the relationship between the rotation angle in the steering wheel control input area 201 and the actual steering wheel angle. For example, drawing a 90° arc in the steering wheel control input area 201 can simulate the actual steering wheel rotating only 10°. Similarly, the relationship between the number of clicks or the duration and force applied in the brake pedal control input area 203 and the simulated actual door pedal opening needs to be calibrated.

[0048] Step 3: Acquire an image of the road ahead, load the simulated vehicle and initial driving parameters, receive current location information, and establish a realistic road environment.

[0049] In this embodiment, step 3 includes:

[0050] Step 301: Identify road objects in the road image, including static road objects and dynamic road objects.

[0051] The static road objects include lane lines, curbs, green belts, guardrails, trees, streetlights, traffic lights, etc., while the dynamic road objects include vehicles, pedestrians, etc.

[0052] Step 302: Simplify the simulated vehicle and the road object according to preset rules.

[0053] In this embodiment, the preset rules include: dividing both sides of the driving lane into several convex quadrilaterals, simplifying the simulated vehicle and other vehicles in the driving lane into rectangles according to proportions, simplifying trees and pedestrians into points, and simplifying guardrails and other isolation facilities into line segments.

[0054] Step 303: Load the simplified road object and its attributes, load the simulated vehicle and initial driving parameters, receive the current location information, and establish a realistic road environment.

[0055] The road object attributes include the road object's coordinates and whether it is accessible.

[0056] The simulated road environment can be created either through local recognition or by uploading the road image to a server, where the server creates the simulated road environment and then distributes it.

[0057] The initial driving parameters include the simulated vehicle's current speed and acceleration curve.

[0058] Step 4: Obtain and save the input information of the preset control operation.

[0059] The input information refers to various preset control operations performed on the simulated driving information input terminal, such as turning the steering wheel and pressing the accelerator.

[0060] After receiving various user operations, the system changes the position and angle of the simulated vehicle in a realistic road environment, thereby providing a realistic driving experience.

[0061] Step 5: Receive the simulated driving termination command and output the evaluation result based on the saved input information of the preset control operation.

[0062] Example 2

[0063] Unlike Embodiment 1, this embodiment provides a specific method for step 5, including:

[0064] Step 501: Determine whether the simulated vehicle exhibits a first preset behavior. If yes, the evaluation result is determined to be unqualified; otherwise, proceed to the next step. The first preset behavior is contact with a road object whose road object attribute is untouchable.

[0065] Step 502: Calculate the acceleration a and the change Δa within the first preset range, and determine whether the change Δa of acceleration is greater than the threshold. If it is, the evaluation result is judged as unqualified; otherwise, proceed to the next step. The first preset range is the preset distance before and after the traffic light and the pedestrian crossing.

[0066] In this embodiment, the first preset range is 20 to 100 meters.

[0067] Step 503: Calculate fuel consumption F based on the acceleration a.

[0068] In a feasible example, when acceleration a ≤ 0, fuel consumption F = 1; when 0 < acceleration a < 1.2, fuel consumption F = 4.17a + 1; when acceleration a > 1.2, fuel consumption F = 7.

[0069] Step 504: Calculate the comprehensive evaluation value P based on the acceleration change Δa and fuel consumption F.

[0070] In this embodiment, P = k1 / Δa + k2*F, where P represents the comprehensive evaluation value, and k1 and k2 are empirical coefficients that can be obtained through big data analysis.

[0071] Step 505: Determine the rating level based on the comprehensive evaluation value P.

[0072] The specific relationship between the overall evaluation score and the rating level can be set according to the actual situation. For example, an overall evaluation score greater than 80 points can be determined as an excellent level, an overall evaluation score between 60 and 80 points can be determined as a general level, and an overall evaluation score less than 60 points can be determined as an unqualified level.

[0073] Example 3

[0074] Unlike Embodiment 2, this embodiment provides a specific method for determining whether the simulated vehicle exhibits a first preset behavior in step 501, including:

[0075] Step A1: Obtain the origin of the simulated vehicle and the i-th road object, as well as the coordinates of each vertex in the world coordinate system.

[0076] The simulated vehicles and road objects are all based on the simplified simulated vehicles and road objects in step 302.

[0077] Step A2: Calculate the distance Di between the origin of the simulated vehicle and the center of the i-th road object.

[0078] Step A3: Calculate the maximum collision detection radius R0(max) and minimum collision detection radius R0(min) of the simulated vehicle, and calculate the maximum collision detection radius Ri(max) and minimum collision detection radius Ri(min) of the i-th road object.

[0079] The collision detection radius refers to the distance between the center point of the object (simulated vehicle and road object) and each edge of the object. The maximum collision detection radius is the largest of these, and the minimum collision detection radius is the smallest of these.

[0080] Step A4: Determine whether Di≤R0(max)+Ri(max) is true. If yes, proceed to the next step; otherwise, determine that no collision has occurred.

[0081] Step A5: Determine whether Di≤R0(min)+Ri(min) is true. If yes, it means that the two have collided; otherwise, proceed to the next step.

[0082] Step A6: Determine whether the simulated vehicle or the i-th road object has at least one vertex within the bounding box of the other. If so, determine that a collision has occurred; otherwise, determine that no collision has occurred.

[0083] Example 4

[0084] Unlike Embodiment 3, this embodiment provides a specific method for determining whether the simulated vehicle or the i-th road object has at least one vertex within the bounding box of the other side in step A6, including:

[0085] Step A6-1: Take any vertex of the simulated vehicle or the i-th road object, and denote it as test vertex P(k);

[0086] Step A6-2: Represent the angle formed by the test vertex P(k) and the two adjacent vertices of the other side as the interior angle ∠Q(m)P(k)Q(m+1), where m=1,2,3…n-1 (n represents the number of vertices of the other side's border);

[0087] Step A6-3: Calculate the sum J of all the interior angles;

[0088] Step A6-4: Determine whether the sum of the angles J is equal to 360°. If so, determine that the test vertex P(k) is located within the border of the other side.

[0089] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A simulated driving evaluation method, characterized in that, include: Step 1: Set the preset control operation information input area on the simulated driving information input terminal; Step 2: Calibrate the correspondence between the preset control operations and the actual vehicle control operations; Step 3: Acquire an image of the road ahead, load the simulated vehicle and initial driving parameters, receive current location information, and establish a realistic road environment; Step 4: Obtain and save the input information for the preset control operation; Step 5: Receive the simulated driving termination command, and output the evaluation result based on the saved input information of the preset control operation; Step 3 includes: Step 301: Identify road objects in the road image, including static road objects and dynamic road objects; Step 302: Simplify the simulated vehicle and the road object according to preset rules; Step 303: Load the simplified road object and its attributes, load the simulated vehicle and initial driving parameters, receive the current location information, and establish a realistic road environment; Step 5 includes: Step 501: Determine whether the simulated vehicle has a first preset behavior. If yes, the evaluation result is judged as unqualified. Otherwise, proceed to the next step. The first preset behavior is contact with a road object whose road object attribute is untouchable. Step 502: Calculate the acceleration and change within the first preset range, and determine whether the change in acceleration is greater than the threshold. If it is, the evaluation result is determined to be unqualified; otherwise, proceed to the next step. The first preset range is the preset distance before and after the traffic light and the pedestrian crossing. Step 503: Calculate fuel consumption based on the acceleration; Step 504: Calculate the comprehensive evaluation value based on the acceleration change and fuel consumption.

2. The simulated driving evaluation method according to claim 1, characterized in that, The first preset range is 20 to 100 meters.

3. The simulated driving evaluation method according to claim 1, characterized in that, Step 503 includes: when acceleration a≤0, fuel consumption F=1; when 0<acceleration a<1.2, fuel consumption F=4.17a+1; when acceleration a>1.2, fuel consumption F=7.

4. The simulated driving evaluation method according to claim 3, characterized in that, The comprehensive evaluation value is calculated according to the following formula: P=k1 / Δa+k2*F, where P represents the comprehensive evaluation value, and k1 and k2 are empirical coefficients.

5. The simulated driving evaluation method according to claim 1, characterized in that, The step 501 of determining whether the simulated vehicle exhibits the first preset behavior includes: Step A1: Obtain the origin and coordinates of each vertex of the simulated vehicle and the i-th road object in the world coordinate system; Step A2: Calculate the distance Di between the origin of the simulated vehicle and the center of the i-th road object; Step A3: Calculate the maximum collision detection radius R0(max) and minimum collision detection radius R0(min) of the simulated vehicle, and calculate the maximum collision detection radius Ri(max) and minimum collision detection radius Ri(min) of the i-th road object; Step A4: Determine whether Di≤R0(max)+Ri(max) is true. If yes, proceed to the next step; otherwise, determine that no collision has occurred. Step A5: Determine whether Di≤R0(min)+Ri(min) is true. If yes, it means that the two have collided; otherwise, proceed to the next step. Step A6: Determine whether the simulated vehicle or the i-th road object has at least one vertex within the bounding box of the other. If so, determine that a collision has occurred; otherwise, determine that no collision has occurred.

6. The simulated driving evaluation method according to claim 5, characterized in that, Step A6, determining whether the simulated vehicle or the i-th road object has at least one vertex within the bounding box of the other side, includes: Step A6-1: Take any vertex of the simulated vehicle or the i-th road object, and denote it as test vertex P(k); Step A6-2: Represent the angle formed by the test vertex P(k) and the two adjacent vertices of the other side as the interior angle ∠Q(m)P(k)Q(m+1), where m=1,2,3…n-1 (n represents the number of vertices of the other side's border); Step A6-3: Calculate the sum J of all the interior angles; Step A6-4: Determine whether the sum of the angles J is equal to 360°. If so, determine that the test vertex P(k) is located within the border of the other side.

Images