A method and terminal for intelligent selection of following vehicles for autonomous vehicles

Abstract

This invention discloses a method and terminal for intelligently selecting following vehicles for autonomous vehicles. The method involves detecting the number of lanes and the behavior of vehicle segments within a preset line-of-sight in each lane; scoring the vehicle's driving stability based on the vehicle segment behavior to identify vehicles whose scores meet a preset threshold; and selecting any vehicle whose score meets the preset threshold and maintaining a safe distance while following it. This invention utilizes the autonomous vehicle's monitoring system to acquire the number of lanes and the behavior of vehicle segments within a preset line-of-sight in each lane. It analyzes the vehicle segment behavior in each lane to score the vehicle's driving stability, filtering out unstable vehicles, such as those exhibiting frequent lane changes, overtaking, or emergency braking, obtaining data on vehicles driving normally, and selecting one to follow at a safe distance. This method improves the safety of autonomous driving by finding suitable following vehicles in complex road conditions.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving, and in particular to a method and terminal for an autonomous vehicle to intelligently select following vehicles. Background Technology

[0002] Currently, with the continuous development of autonomous driving technology, the application of autonomous vehicles on roads has become a reality. However, in the process of autonomous driving, it is impossible to drive automatically without other vehicles. It also needs to deal with complex and diverse road conditions. When faced with multiple lanes and multiple vehicles, the decision-making problem of choosing which vehicle to follow has become a technical problem that urgently needs to be solved.

[0003] In traditional autonomous driving systems, vehicles typically select the vehicle in front as the follower. However, this simple selection strategy is not suitable for complex road conditions with multiple lanes and multiple vehicles. Sometimes, the vehicle in front may be engaging in unstable behaviors such as frequent lane changes, overtaking, or emergency braking, which can pose safety hazards to autonomous vehicles behind. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and terminal for intelligent selection of following vehicles for autonomous vehicles, so as to solve the problem of difficulty in deciding which vehicle to follow when facing multiple lanes and multiple vehicles in the process of autonomous driving.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A method for intelligently selecting following vehicles for an autonomous vehicle includes the following steps:

[0007] S1, detects the number of lanes and the behavior of vehicles within a preset line of sight in each lane;

[0008] S2. The vehicle driving stability is scored based on the vehicle's behavior on the road segment to obtain a vehicle whose score meets expectations;

[0009] S3. Select any vehicle whose rating meets expectations and maintain a safe distance while following it.

[0010] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0011] A terminal for intelligent selection of following vehicles in an autonomous vehicle includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it completes the steps of a method for intelligent selection of following vehicles in an autonomous vehicle.

[0012] The beneficial effects of this invention are as follows: It provides a method and terminal for intelligently selecting following vehicles for autonomous vehicles. The method utilizes the monitoring system of the autonomous vehicle to obtain the number of lanes and the behavior of vehicle segments within a preset line of sight in each lane. Based on the analysis of the vehicle segment behavior in each lane, the driving stability of the vehicle is scored, and unstable vehicles, such as those that frequently change lanes, overtake, or brake suddenly, are filtered out. Data on vehicles that are driving normally is obtained, and one of them is selected to follow at a safe distance. This method finds suitable following vehicles in complex road conditions and improves the safety of autonomous driving. Attached Figure Description

[0013] Figure 1 This is a flowchart of a method for intelligent selection of following vehicles by an autonomous vehicle according to Embodiment 1 of the present invention;

[0014] Figure 2 This is a flowchart of vehicle scoring in a method for intelligent selection of following vehicles by an autonomous vehicle according to Embodiment 4 of the present invention;

[0015] Figure 3 This is a flowchart illustrating a method for intelligent selection of following vehicles by an autonomous vehicle in Embodiment 4 of the present invention.

[0016] Figure 4 This is a schematic diagram of road condition acquisition for a method of intelligent selection of following vehicles by an autonomous vehicle in Embodiment 4 of the present invention;

[0017] Figure 5 This is a schematic diagram of a terminal for intelligent selection of following vehicles in an autonomous vehicle according to Embodiment 4 of the present invention;

[0018] Label Explanation:

[0019] 1. A terminal for intelligently selecting following vehicles in an autonomous vehicle; 2. A memory;

[0020] 3. Processor. Detailed Implementation

[0021] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0022] Please refer to Figure 1 and Figure 4 A method for an autonomous vehicle to intelligently select following vehicles includes the following steps:

[0023] S1, detects the number of lanes and the behavior of vehicles within a preset line of sight in each lane;

[0024] S2. The vehicle driving stability is scored based on the vehicle's behavior on the road segment to obtain a vehicle whose score meets expectations;

[0025] S3. Select any vehicle whose rating meets expectations and maintain a safe distance while following it.

[0026] As can be seen from the above description, the beneficial effects of the present invention are as follows: by using the monitoring system of autonomous vehicles to obtain the number of lanes and the behavior of vehicle segments within a preset line of sight in each lane, and by analyzing the behavior of vehicle segments in each lane, the driving stability of the vehicle is scored, unstable vehicles, such as vehicles that frequently change lanes, overtake, or brake suddenly, are filtered out, data of vehicles driving normally are obtained, and one of them is selected to follow at a safe distance, thus finding a suitable vehicle to follow in complex road conditions and improving the safety of autonomous driving.

[0027] Please refer to Figure 2 In an embodiment of the present invention, step S2 specifically includes the following steps:

[0028] S21. Analyze the road segment behavior of each vehicle and record the abnormal driving behavior of each vehicle within the first preset period.

[0029] S22. Accumulate scores based on preset weighted scores of the abnormal driving behavior, and filter out vehicles with scores below a preset threshold.

[0030] As described above, abnormal driving behaviors of vehicles are recorded within the first preset period, and a preset weight score for each abnormal driving behavior is accumulated to obtain a score for each vehicle. Vehicles with scores below a preset threshold are selected, meaning that vehicles with scores below the preset threshold are considered to be driving normally and have higher following safety.

[0031] Specifically, the abnormal driving behaviors include: lane changing, overtaking, emergency braking, and activating hazard lights. Correspondingly, each abnormal driving behavior has a preset weight score of two, three, four, and five points, respectively, and a preset threshold of eight points. For example, if a vehicle overtakes and activates its hazard lights within the first preset period, the vehicle score is eight points, which is equal to the preset threshold. At this point, following this vehicle is considered unsafe, and therefore, following the vehicle is not considered.

[0032] Preferably, after selecting a vehicle for following, the cumulative score of vehicles within a preset line of sight is detected and calculated cyclically within a first preset period. When it is found that the cumulative score of the current vehicle is no longer lower than a preset threshold, the following vehicle is switched in a timely manner.

[0033] In an embodiment of the present invention, step S22 specifically comprises:

[0034] The abnormal driving behavior is scored by accumulating a score based on a preset weight score. Vehicles with scores greater than or equal to a preset threshold are marked as special vehicles, and vehicles with scores lower than the preset threshold are filtered out.

[0035] If all vehicles within the preset line of sight in each lane are special vehicles, then follow any vehicle at the first preset distance until a vehicle with a score lower than the preset threshold appears within the preset line of sight, and proceed to step S3.

[0036] As can be seen from the above description, during the autonomous driving process, it is inevitable that all vehicles within the preset line of sight will be special vehicles (vehicles with a score greater than or equal to the preset threshold). When this situation occurs, the vehicle will follow any vehicle at a first preset distance. Specifically, the first preset distance is set according to the road conditions and vehicle type, and is 1.5 to 2 times the normal following distance to ensure safety during the autonomous driving process.

[0037] Please refer to Figure 3 In an embodiment of the present invention, step S3 specifically includes the following steps:

[0038] S31. Select any vehicle whose score meets the preset threshold and follow it while maintaining the second preset distance;

[0039] S32. Calculate the vehicle's speed and analyze to determine the optimal following distance;

[0040] S33. Maintain the optimal following distance for the second preset time period and evaluate the speed stability of the vehicle. If the stability is good, proceed to step S34; otherwise, proceed to step S35.

[0041] S34. Follow the vehicle at the same speed as it is traveling;

[0042] S35. Maintain the optimal following distance when following the vehicle.

[0043] As described above, after identifying the vehicle to follow, the system first maintains a second preset close following distance. This involves driving to a distance of the second preset distance from the vehicle in front to facilitate data analysis and subsequent adjustments. Preferably, the second preset distance is set based on road conditions. Then, the optimal following distance is calculated based on the speed of the vehicle in front and maintained at this optimal distance for a second preset duration. Subsequently, the speed stability of the vehicle in front is evaluated within the second preset duration. If the stability is good, the system maintains the same speed as the vehicle in front; otherwise, the optimal following distance is maintained. Specifically, the second preset duration is a safety time interval, ranging from 2 to 3 seconds.

[0044] Specifically, step S32 includes the following steps:

[0045] S321. Calculate the vehicle's speed and, based on that speed, calculate the minimum following distance.

[0046] S322. The optimal following distance is calculated using the following formula:

[0047] D = v1 * T + D min ;

[0048] In the formula:

[0049] D represents the optimal following distance;

[0050] v1 represents the vehicle's speed;

[0051] T represents the third preset duration;

[0052] D min This represents the minimum following distance.

[0053] As can be seen from the above description, within the security event interval

[0054] In an embodiment of the present invention, step S3 is followed by step S4:

[0055] The vehicle's behavior on the road segment is monitored in real time. When the vehicle being followed exhibits abnormal driving behavior, the process returns to step S1.

[0056] As can be seen from the above description, the minimum following distance is calculated based on the speed of the vehicle in front, and a safe interval is reserved to ensure that the vehicle in front has enough time and space to handle the emergency if the vehicle in front encounters a problem.

[0057] In an embodiment of the present invention, step S20 is included before step S2:

[0058] Identify and filter out large vehicles within a preset line of sight in the lane, and abandon the option to follow these large vehicles.

[0059] As described above, due to the large size of large vehicles, there are numerous blind spots when following them, affecting the vehicle's automatic navigation function's judgment of road conditions ahead. Therefore, when the vehicle in front is a large vehicle, the system automatically abandons following. It should be noted that large vehicles, as commonly understood, are various motor vehicles with a gross vehicle weight of 4.5 tons or more, or with 9 or more passenger seats.

[0060] In an embodiment of the present invention, step S1 specifically comprises:

[0061] The system detects the number of lanes and the behavior of vehicles within a preset line of sight in each lane. If there are no vehicles in any lane, the system proceeds at a preset speed; otherwise, it proceeds to step S2.

[0062] As can be seen from the above description, during the autonomous driving process, it is inevitable that there will be situations where there are no cars to follow in the surrounding lanes. At this time, the vehicle is controlled to drive at a preset speed. Preferably, when the vehicle starts to drive, the road information system is automatically imported. The vehicle road information is imported into the intelligent driving vehicle system through the satellite system, and the preset speed is determined based on the road information and vehicle information.

[0063] Please refer to Figure 5 A terminal 1 for intelligent selection of following vehicles by an autonomous vehicle includes a memory 2, a processor 3, and a computer program stored in the memory 2 and executable on the processor. When the processor 3 executes the computer program, it completes the steps in a method for intelligent selection of following vehicles by an autonomous vehicle.

[0064] As described above, an execution platform for a method of intelligently selecting following vehicles for autonomous vehicles is provided. When executing the above method, the monitoring system of the autonomous vehicle is used to obtain the number of lanes and the behavior of vehicle segments within a preset line of sight in each lane. The vehicle driving stability is scored based on the behavior of vehicle segments in each lane, and unstable vehicles, such as those that frequently change lanes, overtake, or brake suddenly, are filtered out. Data on vehicles driving normally is obtained, and one of them is selected to follow at a safe distance. This method finds suitable following vehicles in complex road conditions and improves the safety of autonomous driving.

[0065] The method and terminal for intelligent selection of following vehicles for autonomous vehicles provided by the present invention are mainly used to select suitable vehicles for following during autonomous driving. The following is a detailed description with reference to the embodiments.

[0066] Please refer to Figure 1 Embodiment 1 of the present invention is as follows:

[0067] A method for intelligently selecting following vehicles for an autonomous vehicle includes the following steps:

[0068] S1, detects the number of lanes and the behavior of vehicles within a preset line of sight in each lane;

[0069] S2. The vehicle driving stability is scored based on the vehicle's behavior on the road segment to obtain a vehicle whose score meets expectations;

[0070] S3. Select any vehicle whose rating meets expectations and maintain a safe distance while following it.

[0071] In this embodiment, the monitoring system of the autonomous vehicle is used to obtain the number of lanes and the behavior of vehicle segments within a preset line of sight in each lane. The vehicle driving stability is scored based on the behavior of vehicle segments in each lane, and unstable vehicles, such as those that frequently change lanes, overtake, or brake suddenly, are filtered out. Data on vehicles that are driving normally is obtained, and one of them is selected to follow at a safe distance. In complex road conditions, a suitable vehicle to follow is found, thereby improving the safety of autonomous driving.

[0072] Please refer to Figure 2 Embodiment two of the present invention is as follows:

[0073] Based on Example 1, step S2 specifically includes the following steps:

[0074] S21. Analyze the road segment behavior of each vehicle and record the abnormal driving behavior of each vehicle within the first preset period. Abnormal driving behavior includes: lane changing, overtaking, emergency braking and activating hazard lights. Correspondingly, the preset weight scores for each abnormal driving behavior are two, three, four and five points, respectively, and the preset threshold is eight points.

[0075] S22. If all vehicles within the preset line-of-sight distance in each lane are special vehicles, then follow any vehicle at a first preset distance until a vehicle with a score below a preset threshold appears within the preset line-of-sight distance, and proceed to step S3. Abnormal driving behaviors of vehicles are recorded within a first preset period, and a preset weighted score for each abnormal driving behavior is accumulated to obtain a final score for each vehicle. Vehicles with scores below the preset threshold are selected, meaning they are considered normal driving vehicles with higher following safety.

[0076] In addition, during the autonomous driving process, it is inevitable that some vehicles within the preset line of sight will be special vehicles (vehicles with a score greater than or equal to the preset threshold). When this happens, the vehicle will follow any vehicle at a first preset distance. Specifically, the first preset distance is set according to road conditions and vehicle type, and is 1.5 to 2 times the normal following distance to ensure safety during the autonomous driving process.

[0077] Preferably, after selecting a vehicle for following, the cumulative score of vehicles within a preset line of sight is detected and calculated cyclically within a first preset period. When it is found that the cumulative score of the current vehicle is no longer lower than a preset threshold, the following vehicle is switched in a timely manner.

[0078] Please refer to Figure 3 Embodiment 3 of the present invention is as follows:

[0079] Based on Example 1, step S3 specifically includes the following steps:

[0080] S31. Select any vehicle whose score meets the preset threshold and follow it while maintaining the second preset distance. After determining the vehicle to follow, first maintain the second preset close distance while following it, that is, first drive to the second preset distance from the vehicle in front, so as to facilitate the statistical calculation of the data of the vehicle in front and facilitate subsequent adjustments.

[0081] S32. Calculate the vehicle's speed and analyze to determine the optimal following distance;

[0082] S33. Maintain the optimal following distance for the second preset time period and evaluate the speed stability of the vehicle. If the stability is good, proceed to step S34; otherwise, proceed to step S35. Specifically, the second preset time period is a safe time interval, with a value between 2 and 3 seconds.

[0083] S34. Follow the vehicle at the same speed as it is traveling;

[0084] S35. Maintain the optimal following distance when following the vehicle.

[0085] Step S32 specifically includes the following steps:

[0086] S321. Calculate the vehicle's speed and, based on that speed, calculate the minimum following distance.

[0087] S322. The optimal following distance is calculated using the following formula:

[0088] D = v1 * T + D min ;

[0089] In the formula:

[0090] D represents the optimal following distance;

[0091] v1 represents the vehicle's speed;

[0092] T represents the third preset duration;

[0093] D min This represents the minimum following distance.

[0094] Embodiment four of the present invention is as follows:

[0095] Based on Example 1, step S3 is followed by step S4:

[0096] The system monitors the vehicle's behavior on the road segment in real time. When the followed vehicle exhibits abnormal driving behavior, it returns to step S1. The minimum following distance is calculated based on the speed of the vehicle in front, and a safe zone is reserved to ensure that the vehicle has sufficient time and space to handle emergencies if the vehicle in front encounters a problem.

[0097] Step S20 precedes step S2:

[0098] The system identifies and filters out large vehicles within a preset line of sight in the lane, and abandons the option to follow large vehicles. Because large vehicles are large, they create significant blind spots when following them, affecting the vehicle's automatic navigation system's assessment of road conditions ahead. Therefore, when the vehicle in front is a large vehicle, the system automatically abandons the option to follow it. It should be noted that large vehicles are generally understood to be any motor vehicle with a gross vehicle weight of 4.5 tons or greater, or with 9 or more passenger seats.

[0099] Please refer to Figures 2 to 4 Embodiment four of the present invention is as follows: A specific application of a method for intelligently selecting following vehicles for autonomous vehicles is as follows:

[0100] (1) When the autonomous vehicle travels to a three-lane road, it first imports the road information system to obtain lane information and speed limit requirements, detects the vehicle situation within the preset line of sight, and makes a preliminary judgment on whether there are vehicles on the road. If there are no vehicles in all lanes ahead, it travels at the preset speed and keeps monitoring the road conditions.

[0101] (2) If there are vehicles in the lane within sight distance, the driving stability score is calculated based on the behavior of the vehicles in the lane. If a large vehicle is found during the process, no score is required, the vehicle will automatically give up following and the other vehicles will be scored.

[0102] The scoring process is as follows: The road behavior of each vehicle is analyzed, and abnormal driving behaviors of each vehicle are recorded within a first preset period. Abnormal driving behaviors include lane changing, overtaking, emergency braking, and activating hazard lights. Correspondingly, each abnormal driving behavior has a preset weight score of two, three, four, and five points, with a preset threshold of eight points. If a vehicle's preset behavior within the first preset period is below the preset threshold, it is considered safe to follow; otherwise, it is marked as a special vehicle. If all vehicles within a preset line of sight are special vehicles, then any vehicle is cautiously followed at a first preset distance. Simultaneously, if a label indicating a novice driver is detected on a vehicle ahead, one point is automatically added to the scoring process. Furthermore, after selecting a vehicle to follow, the cumulative score of vehicles within a preset line of sight is continuously detected and calculated within the first preset period. When the current vehicle's cumulative score is no longer below the preset threshold, the vehicle to be followed is switched promptly.

[0103] (3) Select any vehicle with a score lower than the preset threshold for following, and first maintain a second preset distance S from the vehicle. q During the driving process, the speed of the vehicle in front, v1, is calculated and the speed of the vehicle in front, v2, and the optimal following distance is obtained.

[0104] Within a safe time interval (a second preset duration, ranging from 2 to 3 seconds, preferably 3 seconds), follow the vehicle ahead at the optimal following distance and evaluate the speed stability of the vehicle ahead. If the speed stability of the vehicle ahead is good, maintain the same speed as the vehicle ahead (i.e., ensure consistent speed). If the speed of the vehicle ahead is unstable during the same speed following period, maintain the optimal following distance. If the speed stability of the vehicle ahead is average or below, maintain the optimal following distance (i.e., ensure constant distance).

[0105] If the vehicle in front exhibits abnormal driving behavior while following another vehicle, follow the steps above to select a new vehicle to follow.

[0106] The optimal following distance is calculated using the following formula:

[0107] D = v1 * T + D min ;

[0108] In the formula:

[0109] D represents the optimal following distance;

[0110] v1 represents the vehicle's speed;

[0111] T represents the third preset duration;

[0112] D min This represents the minimum following distance.

[0113] Please refer to Figure 5 Embodiment 5 of the present invention is: a terminal 1 for intelligent selection of following vehicles by an autonomous vehicle, including a memory 2, a processor 3, and a computer program stored on the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, it completes the steps in any of the methods for intelligent selection of following vehicles by an autonomous vehicle in Embodiments 1 to 4.

[0114] In summary, the present invention provides a method and terminal for intelligent selection of following vehicles for autonomous vehicles. It utilizes the autonomous vehicle's monitoring system to acquire the number of lanes and the behavior of vehicle segments within a preset line-of-sight in each lane. Based on the analysis of the vehicle segment behavior in each lane, the driving stability of the vehicle is scored, and unstable vehicles, such as those exhibiting frequent lane changes, overtaking, or emergency braking, are filtered out. Data on vehicles driving normally is obtained, and one of these vehicles is selected to follow at a safe distance. This method finds suitable following vehicles in complex road conditions, thereby improving the safety of autonomous driving.

[0115] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for intelligently selecting following vehicles for an autonomous vehicle, characterized in that: Including the following steps: S1, detects the number of lanes and the behavior of vehicles within a preset line of sight in each lane; S2. The vehicle driving stability is scored based on the vehicle's behavior on the road segment to obtain a vehicle whose score meets expectations; Step S2 specifically includes the following steps: S21. Analyze the road segment behavior of each vehicle and record the abnormal driving behavior of each vehicle within the first preset period. The abnormal driving behaviors include: changing lanes, overtaking, emergency braking, and activating hazard lights. S22. Accumulate scores based on preset weighted scores of the abnormal driving behavior, and filter out vehicles with scores below a preset threshold. S3. Select any vehicle whose rating meets expectations and maintain a safe distance while following it. Step S3 specifically includes the following steps: S31. Select any vehicle whose score meets the preset threshold and follow it while maintaining the second preset distance; S32. Calculate the vehicle's speed and analyze to determine the optimal following distance; Step S32 specifically includes the following steps: S321. Calculate the vehicle's speed and, based on that speed, calculate the minimum following distance. S322. The optimal following distance is calculated using the following formula: ； In the formula: D represents the optimal following distance; v1 represents the vehicle's speed; T represents the third preset duration; D min This represents the minimum following distance. S33. Maintain the optimal following distance for the second preset time period and evaluate the speed stability of the vehicle. If the stability is good, proceed to step S34; otherwise, proceed to step S35. S34. Follow the vehicle at the same speed as it is traveling; S35. Maintain the optimal following distance when following the vehicle.

2. The method for intelligently selecting following vehicles for an autonomous vehicle according to claim 1, characterized in that: Step S22 specifically involves: The abnormal driving behavior is scored by accumulating a score based on a preset weight score. Vehicles with scores greater than or equal to a preset threshold are marked as special vehicles, and vehicles with scores lower than the preset threshold are filtered out. If all vehicles within the preset line of sight in each lane are special vehicles, then follow any vehicle at the first preset distance until a vehicle with a score lower than the preset threshold appears within the preset line of sight, and proceed to step S3.

3. A method for intelligently selecting following vehicles for an autonomous vehicle according to claim 1 or 2, characterized in that: Step S3 is followed by step S4: The vehicle's behavior on the road segment is monitored in real time. When the vehicle being followed exhibits abnormal driving behavior, the process returns to step S1.

4. The method for intelligently selecting following vehicles for an autonomous vehicle according to claim 1, characterized in that: Step S20 is included before step S2: Identify and filter out large vehicles within a preset line of sight in the lane, and abandon the option to follow these large vehicles.

5. The method for intelligently selecting following vehicles for an autonomous vehicle according to claim 1, characterized in that: Step S1 specifically involves: The system detects the number of lanes and the behavior of vehicles within a preset line of sight in each lane. If there are no vehicles in any lane, the system travels at a preset speed. Otherwise, proceed to step S2.

6. A terminal for intelligently selecting following vehicles in an autonomous vehicle, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the steps of the method for intelligent selection of following vehicles by an autonomous vehicle as described in any one of claims 1-5.

Images