A lane change decision system and method based on logical rules

Abstract

The application is suitable for the field of automatic driving technology, and provides a lane changing decision system and method based on logic rules, which comprises: a sensor information preprocessing module; a lane condition judgment module for judging lane conditions; a lane changing demand judgment module for judging lane changing demands; a lane changing safety judgment module for judging lane changing safety; and a lane changing decision module for comprehensively deciding the judgment results of lane conditions, lane changing demands and lane changing safety. The application establishes a lane changing decision model through a Stateflow hierarchical state machine model, verifies that the lane changing decision algorithm can realize lane changing decision of vehicles on straight roads in urban areas through the lane changing decision model, greatly reduces traffic accidents caused by lane changing, effectively guarantees the safety and reliability of lane changing of vehicles in complex environments, reduces the driving pressure of drivers, and improves driving comfort.

Description

Technical Field

[0001] This invention belongs to the field of autonomous driving technology, and in particular relates to a lane-changing decision system and method based on logical rules. Background Technology

[0002] With the rapid development of the global economy and technology, the number of cars in countries around the world has increased rapidly. While the increase in the number of cars has brought great convenience to people's daily travel, it has also brought serious challenges to traffic safety and road traffic. In order to ensure driver safety and improve the public's travel environment, and with the rapid development of Internet technology and artificial intelligence technology, autonomous driving technology has become an important development and research direction for the global automotive industry. In recent years, major automakers have carried out research on autonomous driving technology. Foreign automakers have developed many Level 2 autonomous driving systems, such as ACC, CCS, and AEB systems, to assist drivers, reduce driver stress, and ensure driving safety. Domestic automakers have also launched Level 2 autonomous driving vehicles, such as the Geely Binyue and Great Wall Haval F7, which are already in mass production. It can be seen that Level 2 autonomous driving technology at home and abroad is approaching maturity, and Level 3 autonomous driving will become the next focus of technological competition.

[0003] Autonomous driving technology mainly consists of four modules: environmental perception, behavior decision-making, trajectory planning, and control execution. During the driving process, the vehicle collects information about the surrounding road environment through onboard cameras, radar, and other sensors for data processing. The processed data is then transmitted to the onboard terminal for behavior decision-making. Based on the road environment, the onboard terminal plans the vehicle's driving path, and finally, the actuator controls the vehicle to drive along the expected path.

[0004] The behavior decision-making module plays a decisive role in autonomous driving systems. Lane changing is a fundamental action necessary for autonomous vehicles navigating multi-lane roads, and it is also essential for achieving highly automated driving. However, due to the influence of complex environments and errors in micro-data acquisition, lane changing decision algorithms still require further improvement. Based on current research, lane changing decision technology for autonomous vehicles can be considered to fall into two main research directions: logic rule-based decision-making and machine learning-based decision-making. This application optimizes logic rule-based decision-making to address the shortcomings in autonomous lane changing decision-making algorithms for autonomous vehicles. Summary of the Invention

[0005] The purpose of this invention is to provide a lane-changing decision system and method based on logical rules, which aims to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A lane-changing decision-making system based on logical rules includes: a lane condition judgment module, a lane-changing demand judgment module, a lane-changing safety judgment module, a lane-changing decision module, and a sensor information preprocessing module;

[0008] The lane condition determination module is used to determine lane conditions;

[0009] The lane change requirement determination module is used to determine lane change requirements;

[0010] The lane change safety assessment module is used to assess lane change safety.

[0011] The lane-changing decision module is used to make a comprehensive decision based on the judgment results of lane conditions, lane-changing needs, and lane-changing safety.

[0012] Furthermore, the lane condition judgment module judges lane conditions based on the analysis of lane line marker sensor data, including: the Matlab Function module outputs the corresponding left / right lane change markers based on the lane line validity and lane line ID of the analyzed vehicle's left and right lanes.

[0013] Furthermore, the lane-change demand determination module receives current lane vehicle information filtered by the sensor information preprocessing module: relative distance and relative speed. The driver's lane-change demand is affected by the following factors:

[0014] The driver's desired speed is quantified by the vehicle speed dissatisfaction and the cumulative vehicle speed dissatisfaction.

[0015] There is a possibility of collision with the vehicle ahead; the occurrence of lane-changing intention is measured by the TTC collision time.

[0016] Input the final calculated vehicle speed dissatisfaction accumulation and TTC collision time into the Matlab Function module. When either the vehicle speed dissatisfaction accumulation or the TTC collision time is within the threshold range, it is determined that a lane change request has been generated, and the Matlab Function module outputs the corresponding lane change request flag.

[0017] Furthermore, the formula for calculating the vehicle speed non-compliance is as follows:

[0018]

[0019] Where K represents the vehicle speed dissipation over a period of time, and e is the gain coefficient. The current vehicle speed, For the vehicle's desired speed, This is the upper limit of vehicle speed error. The sampling time interval;

[0020] The formula for calculating the cumulative percentage of vehicle speed not being fully accumulated is:

[0021]

[0022] in, The unsatisfactory cumulative rate of vehicle speed after the nth sampling time interval;

[0023] When the vehicle speed is less than the cumulative threshold Km, that is: The intention to change lanes arises;

[0024] The formula for calculating TTC collision time is:

[0025]

[0026] Where S is the distance between the obstacle in front and the vehicle. The speed of the obstacle relative to the vehicle;

[0027] When TTC is within a certain range, that is: The intention to change lanes arises.

[0028] Furthermore, the lane change safety judgment module receives target lane vehicle information filtered by the sensor information preprocessing module: the relative distance and relative speed of the vehicles ahead detected by the front radar sensor, and the relative distance and relative speed of the vehicles behind detected by the rear radar sensor. The lane change safety judgment module plans a lane change safety zone in the target lane to determine the safety of the lane change. The safety zone is represented by the safe distances of the vehicles ahead and behind. The Matlab Function module predicts the safety zones ahead and behind respectively and outputs the corresponding safety markers.

[0029] Furthermore, the formula for calculating the safe distance to the vehicle ahead is:

[0030]

[0031] The formula for calculating the safe distance from vehicles behind is:

[0032]

[0033] in, This indicates the longitudinal distance between the test vehicle and the vehicle in front. This represents the longitudinal distance between the test vehicle and the vehicle behind it, where L represents the length of the test vehicle and D represents the lane-changing safety distance threshold. Indicates the longitudinal speed of the test vehicle. Indicates the longitudinal speed of the vehicle ahead. This indicates the longitudinal speed of the vehicle behind.

[0034] Furthermore, the inputs for the lane change decision module to determine whether to brake include the left / right lane change flag output by the lane condition judgment module, the lane change demand flag output by the lane change demand judgment module, the lane change safety flag output by the lane change safety judgment module, and the TTC collision time. Based on the lane change rules, the lane change decision module outputs corresponding follow, lane change, and braking commands through the inter-level state transition of the state machine.

[0035] A lane-changing decision-making method based on logical rules includes the following steps:

[0036] Determine lane conditions based on sensor data;

[0037] Determine lane-changing needs based on sensor data;

[0038] Determine lane change safety based on sensor data;

[0039] Make a comprehensive decision based on the assessment results of lane conditions, lane changing needs, and lane changing safety;

[0040] Three key lane-changing scenarios were constructed using Prescan, and lane-changing decision models were built in Matlab / Simulink. The lane-changing decision results for each scenario were analyzed to verify the feasibility and reliability of the lane condition judgment module, lane-changing demand judgment module, lane-changing safety judgment module, and lane-changing decision module, thus completing the lane-changing decision process for autonomous vehicles.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] This logic rule-based lane-changing decision system and method establishes a lane-changing decision model through a Stateflow hierarchical state machine model. The lane-changing decision model verifies that the lane-changing decision algorithm can realize lane-changing decisions for vehicles on straight roads in urban areas, significantly reducing traffic accidents caused by lane changes, effectively ensuring the safety and reliability of lane changes in complex environments, reducing driver stress, and improving driving comfort. Attached Figure Description

[0043] Figure 1 This is a structural block diagram of the present invention.

[0044] Figure 2 This is a diagram of the lane-changing decision-making model for autonomous vehicles in this invention.

[0045] Figure 3 This is a diagram of the lane condition determination module in this invention.

[0046] Figure 4 This is a diagram of the lane change demand judgment model in this invention.

[0047] Figure 5 This is a diagram of the vehicle speed non-accumulation calculation module in this invention.

[0048] Figure 6 This is a diagram of the lane change safety judgment module in this invention.

[0049] Figure 7 This is a structural diagram of the lane-changing decision module in this invention.

[0050] Figure 8 This is a scene diagram of Embodiment 1 of the present invention.

[0051] Figure 9 This is a parameter variation diagram for Embodiment 1 of the present invention, wherein, Figure 9 .1 represents the relative distance to the vehicle in front in the current lane; Figure 9 .2 represents the relative speed of the vehicle in front in the current lane; Figure 9 .3 represents the collision time TTC; Figure 9 0.4 represents the cumulative degree of vehicle speed not meeting the standard requirement K; Figure 9 .5 is the output image of the lane change requirement judgment module; Figure 9 .6 is the output image of the lane condition judgment module; Figure 9 .7 is the output image of the lane change safety judgment module; Figure 9 .8 is the output image of the lane change decision module.

[0052] Figure 10 This is a scene diagram of Embodiment 2 of the present invention.

[0053] Figure 11 This is a parameter variation diagram for Embodiment 2 of the present invention, wherein, Figure 11 .1 represents the relative distance to the vehicle in front in the current lane; Figure 11 .2 represents the relative speed of the vehicle in front in the current lane; Figure 11 .3 represents the collision time TTC; Figure 11 0.4 represents the cumulative degree of vehicle speed not meeting the standard requirement K; Figure 11 .5 is the output image of the lane change requirement judgment module; Figure 11 .6 is the output image of the lane condition judgment module; Figure 11 .7 is the output image of the lane change safety judgment module; Figure 11 .8 is the output image of the lane change decision module.

[0054] Figure 12 This is a scene diagram of Embodiment 3 of the present invention.

[0055] Figure 13 This is a parameter variation diagram for Embodiment 3 of the present invention, wherein, Figure 13 .1 represents the relative distance to the vehicle in front in the current lane; Figure 13 .2 represents the relative speed of the vehicle in front in the current lane; Figure 13.3 represents the relative distance to the vehicle behind in the target lane; Figure 13 0.4 represents the relative speed of the vehicle behind in the target lane; Figure 13 0.5 represents the relative distance to the vehicle in front in the target lane; Figure 13 .6 represents the relative speed of the vehicle ahead in the target lane; Figure 13 .7 represents the collision time TTC; Figure 13 0.8 represents the cumulative speed K. Figure 13 .9 is the output image of the lane change requirement judgment module; Figure 13 .10 is the output image of the lane condition judgment module; Figure 13 .11 is the output image of the lane change safety judgment module; Figure 13 .12 is the output image of the lane change decision module. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0057] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0058] An embodiment of the present invention provides a lane-changing decision system based on logical rules, comprising: a lane condition judgment module, a lane-changing demand judgment module, a lane-changing safety judgment module, a lane-changing decision module, and a sensor information preprocessing module;

[0059] The lane condition determination module is used to determine lane conditions;

[0060] The lane change requirement determination module is used to determine lane change requirements;

[0061] The lane change safety assessment module is used to assess lane change safety.

[0062] The lane-changing decision module is used to make a comprehensive decision based on the judgment results of lane conditions, lane-changing needs, and lane-changing safety.

[0063] In a preferred embodiment of the present invention, the lane condition judgment module judges lane conditions based on the analysis of lane line marker sensor data, including: the Matlab Function module outputs corresponding left / right lane change markers based on the analysis of the lane line validity of the left and right lanes of the vehicle and the lane ID of the lane line.

[0064] In this embodiment of the invention, the lane condition determination module is structured as follows: Figure 3As shown, the lane condition judgment module works as follows: When the sensor recognizes that the lane lines on both sides of the vehicle are valid and the lane ID of one of the lane lines is a crossable lane ID, it determines that the vehicle can change lanes to the left / right lane, and the corresponding left / right lane change flag bit - Left_flag / Right_flag outputs 1.

[0065] In a preferred embodiment of the present invention, the lane change demand determination module receives current lane vehicle information filtered by the sensor information preprocessing module: relative distance and relative speed. The driver's lane change demand is affected by the following factors:

[0066] The driver's desired speed is quantified by the vehicle speed dissatisfaction and the cumulative vehicle speed dissatisfaction.

[0067] There is a possibility of collision with the vehicle ahead; the occurrence of lane-changing intention is measured by the TTC collision time.

[0068] Input the final calculated vehicle speed dissatisfaction accumulation and TTC collision time into the Matlab Function module. When either the vehicle speed dissatisfaction accumulation or the TTC collision time is within the threshold range, it is determined that a lane change request has been generated, and the Matlab Function module outputs the corresponding lane change request flag.

[0069] In this embodiment of the invention, the lane-changing requirement module mainly determines the current lane situation of the test vehicle. In real life, a driver's lane-changing intention is mainly affected by the following factors:

[0070] First, the driver's desired speed. When a vehicle is traveling on the road, the driver usually does not need to change lanes because the vehicle's instantaneous speed is too low. The main reason the driver chooses to change lanes is because the vehicle in front is continuously traveling at a low speed, causing the vehicle's speed to be unable to reach the driver's desired speed for a long time. When the low speed continues for a period of time, the driver will intend to change lanes. Here, we introduce speed dissatisfaction level and speed dissatisfaction accumulation level to quantify the driver's lane-changing intention.

[0071] Second, there is a possibility of collision with the vehicle in front. When an autonomous vehicle is driving on the road, if there is an obstacle ahead or the vehicle in front stops due to a malfunction, the autonomous vehicle should choose to change lanes or brake to ensure driving safety.

[0072] In a preferred embodiment of the present invention, the formula for calculating the vehicle speed non-compliance is as follows:

[0073]

[0074] Where K represents the vehicle speed dissipation over a period of time, and e is the gain coefficient. The current vehicle speed, For the vehicle's desired speed, This is the upper limit of vehicle speed error. The sampling time interval;

[0075] The formula for calculating the cumulative percentage of vehicle speed not being fully accumulated is:

[0076]

[0077] in, The unsatisfactory cumulative rate of vehicle speed after the nth sampling time interval;

[0078] When the vehicle speed is less than the cumulative threshold Km, that is: The intention to change lanes arises;

[0079] The formula for calculating TTC collision time is:

[0080]

[0081] Where S is the distance between the obstacle in front and the vehicle. The speed of the obstacle relative to the vehicle;

[0082] When TTC is within a certain range, that is: The intention to change lanes arises.

[0083] In this embodiment of the invention, the lane-changing demand judgment model mainly consists of two parts: the calculation of TTC collision time and the calculation of the vehicle speed non-accumulation K. The TTC collision time is calculated by using the ratio of the absolute value of the relative distance to the absolute value of the relative speed. The calculation of the vehicle speed non-accumulation K uses an iterative module, the specific structure of which is as follows: Figure 11 As shown in Figure 4, the module inputs are the test vehicle speed and the relative speed between the test vehicle and the target vehicle, and the output is the vehicle speed dissatisfaction K. The parameters selected for the vehicle speed dissatisfaction technical formula are: gain coefficient K = 20, upper limit of vehicle speed error = 2 m / s, sampling time interval = 0.05 s, and iteration period = 4 s. The finally calculated cumulative vehicle speed dissatisfaction and TTC collision time are input to the Matlab Function module for further judgment. The TTC collision time threshold is set to 2 s-4 s, and the cumulative vehicle speed dissatisfaction threshold is set to 30. When either variable falls within the threshold range, a lane-changing requirement is determined, and the lane-changing requirement flag2 is output as 1.

[0084] In a preferred embodiment of the present invention, the lane change safety judgment module receives target lane vehicle information filtered by the sensor information preprocessing module: the relative distance and relative speed of the vehicles ahead detected by the front radar sensor, and the relative distance and relative speed of the vehicles behind detected by the rear radar sensor; the lane change safety judgment module plans a lane change safety zone in the target lane to judge the safety of the vehicle lane change. The safety zone is represented by the safe distance of the vehicles ahead and the safe distance of the vehicles behind. The Matlab Function module predicts the safety zone for the vehicles ahead and behind, respectively, and outputs the corresponding safety flags.

[0085] In this embodiment of the invention, the lane-change safety judgment module mainly focuses on judging the target lane scenario for the test vehicle's lane change. When a vehicle needs to change lanes, it is necessary to make a safety judgment on the surrounding environment, especially the safety of the target lane for the test vehicle's lane change, to avoid collisions with other vehicles. Lane-change safety during vehicle operation is mainly affected by the following factors: the test vehicle's speed, the speed of surrounding vehicles, and the relative distance between vehicles.

[0086] Due to limitations in radar sensor placement and accuracy, it's impossible to accurately detect the distances between the test vehicle's four corners and other vehicles. Therefore, using lane-changing safety distance analysis to assess lane-changing safety carries significant uncertainty and potential risks. This paper employs a lane-changing safety zone prediction method to determine the safety of a lane change. While the vehicle is in motion, a lane-changing safety zone is planned within the target lane. Front and rear radar sensors continuously monitor vehicles in the target lane. If no other vehicles are detected within the safety zone during the lane change, the lane change is considered safe.

[0087] In a preferred embodiment of the present invention, the formula for calculating the safe distance to the vehicle ahead is:

[0088]

[0089] The formula for calculating the safe distance from vehicles behind is:

[0090]

[0091] in, This indicates the longitudinal distance between the test vehicle and the vehicle in front. This represents the longitudinal distance between the test vehicle and the vehicle behind it, where L represents the length of the test vehicle and D represents the lane-changing safety distance threshold. Indicates the longitudinal speed of the test vehicle. Indicates the longitudinal speed of the vehicle ahead. This indicates the longitudinal speed of the vehicle behind.

[0092] In this embodiment of the invention, the lane change safety judgment module is based on the lane change safety area prediction principle. The calculation formula is programmed into the Matlab Function module to predict the safety area in front and behind, and output the corresponding safety flag bits. The subsequent Matlab Function module receives the two safety flag bit information and makes the final judgment on lane change safety. Only when both the front and rear safety areas are in a safe state, that is, when both flag bits are 1, the lane change safety flag bit flag3 outputs 1, which is considered as lane change safety.

[0093] In a preferred embodiment of the present invention, the inputs for the lane change decision module to determine whether to brake include the left / right lane change flag output by the lane condition judgment module, the lane change demand flag output by the lane change demand judgment module, the lane change safety flag output by the lane change safety judgment module, and the TTC collision time. The lane change decision module outputs corresponding follow, lane change, and braking commands according to the lane change rules and through the inter-level state transition of the state machine.

[0094] In this embodiment of the invention, the lane-changing rules for vehicles are shown in Table 1. To ensure the driving safety of vehicles, a rule for vehicle braking is added to the rule base so that the vehicle can brake in time when it is too close to an obstacle.

[0095] Table 1 Lane Changing Rules

[0096]

[0097] As shown in Table 1, a left / right lane change decision will only be made when the left / right lane change sign, the lane change request sign, and the lane change safety sign are all 1. If any one of the three lane change signs is 0, the vehicle will not make a lane change decision to ensure driving safety.

[0098] A lane-changing decision-making method based on logical rules includes the following steps:

[0099] Determine lane conditions based on sensor data;

[0100] Determine lane-changing needs based on sensor data;

[0101] Determine lane change safety based on sensor data;

[0102] Make a comprehensive decision based on the assessment results of lane conditions, lane changing needs, and lane changing safety;

[0103] Three key lane-changing scenarios were constructed using Prescan, and lane-changing decision models were built in Matlab / Simulink. The lane-changing decision results for each scenario were analyzed to verify the feasibility and reliability of the lane condition judgment module, lane-changing demand judgment module, lane-changing safety judgment module, and lane-changing decision module, thus completing the lane-changing decision process for autonomous vehicles.

[0104] Lane-changing decision model construction and simulation verification based on Prescan and Simulink co-simulation: Based on lane condition judgment module, lane-changing demand judgment module, lane-changing safety judgment module and lane-changing decision module, an autonomous vehicle lane-changing decision model was built in Simulink in conjunction with Prescan.

[0105] Figure 2 The functions of each module in the numbered list are as follows: Module 1 is the perception system; Module 2 is the sensor information preprocessing module; Module 3 is the lane condition judgment module, which uses ALMS to identify lane lines on both sides of the vehicle to determine lane change; Module 4 is the lane change demand judgment module, which determines whether a lane change is needed by setting TTC and the vehicle speed not being fully accumulated threshold; Module 5 is the lane change safety judgment module, which identifies and analyzes the safe area around the vehicle to determine the safety of lane change and prevent collisions; Module 6 is the lane change decision module, which comprehensively analyzes the judgment results of the above three modules to decide whether to change lanes. Only when the output flags of modules 3, 4, and 5 are all 1, the lane change condition is met and a lane change is allowed; otherwise, a lane change is not allowed.

[0106] Simulation scenario construction and result analysis

[0107] Example 1

[0108] The test vehicle was forced to change lanes due to a disabled vehicle ahead. The scenario setup is as follows: Figure 8 As shown: In the two-lane scenario, the following vehicle is the test vehicle, which travels at a constant speed of 10 m / s along the current lane; the vehicle 200 m away from the test vehicle is the malfunctioning vehicle, which travels at a speed of 10 m / s along the current lane. Due to the malfunction, it slows down until it stops. The malfunctioning vehicle travels a distance of 60 m. The test road length is 300 m and the simulation time is 30 s.

[0109] The specific parameter changes of the lane-changing decision model during the simulation are as follows: Figure 9 As shown in the image, the relative distance between the autonomous vehicle and the vehicle in front in the current lane is as follows: Figure 9 As shown in Figure 1, initially, the vehicle in front and the test vehicle are at the same speed and are outside the radar detection range. The vehicle in front gradually decelerates and stops due to a malfunction, and the relative distance between the vehicles gradually decreases. The test vehicle's front radar detects the vehicle in front in approximately 11 seconds. The relative speed image between the autonomous vehicle and the vehicle in front in the current lane is shown in Figure 1. Figure 9 As shown in Figure 2, the vehicle in front stopped at 12.5 seconds, and the relative speed between the two vehicles remained at -10 m / s. 2 Left and right. The lane change demand judgment module calculates the collision time TTC image based on the above two parameters. Figure 9 .3) and the cumulative degree of vehicle speed K (image) Figure 9 .4).

[0110] During this process, the cumulative speed K never reached the set threshold. The collision time TTC gradually decreased as the relative distance between the two vehicles decreased. The TTC time reached the TTC threshold range at 23.5s, triggering a lane change request. The lane change request flag bit flag2=1 was output, and the output image of the lane change request module is as follows. Figure 9 As shown in Figure 5. At this time, the lane condition judgment module's flag output image is as follows: Figure 9 As shown in Figure 6, the lane markings are clear and valid, and the left lane marking is a crossable lane marking, with the left lane change flag (Left_flag=1). The lane change safety assessment module outputs the flag image as shown. Figure 9 As shown in Figure 7, since there are no other vehicles in the target lane for lane changing, the lane changing safety flag3 = 1. At 23.5s, all three module flags output 1, and the lane changing decision module's lane changing decision flag = 1. Figure 9 .8), execute lane change.

[0111] Example 2

[0112] The test vehicle was prompted to change lanes due to a slow-moving vehicle ahead. The scenario setup is as follows: Figure 10 As shown: In the two-lane scenario, the following vehicle is the test vehicle, which travels at a constant speed of 10 m / s along the current lane; the vehicle in front, 80 m away from the test vehicle, travels at a constant speed of 8.33 m / s along the current lane. The test road is 500 m long and the simulation time is 45 s.

[0113] The specific parameter changes of the lane-changing decision model during the simulation are as follows: Figure 11 As shown in the image, the relative distance between the autonomous vehicle and the vehicle in front in the current lane is as follows: Figure 11 As shown in Figure 1, initially, the vehicle ahead is within the radar detection range. Because the test vehicle's speed is higher than the vehicle ahead, the relative distance between the vehicles gradually decreases; the relative speed image between the autonomous vehicle and the vehicle ahead in the current lane is shown in Figure 1. Figure 11 As shown in Figure 2, the relative speed between the two vehicles remains at -1.7 m / s. 2 Left and right. The lane change demand judgment module calculates the collision time (TTC) and vehicle speed underrun cumulative degree (K) based on the above two parameters, as shown in the image. Figure 11 As shown in .3.

[0114] During this process, the collision time (TTC) gradually decreases as the relative distance between the two vehicles decreases. Before the TTC reaches the TTC threshold, the cumulative speed K reaches the threshold around 36 seconds, triggering a lane-change request. The lane-change request flag (flag2=1) is then output, and the lane-change request judgment module outputs an image as shown below. Figure 11 As shown in Figure 5. At this time, the lane condition judgment module's flag output image is as follows: Figure 11As shown in Figure 6, the lane markings are clear and valid, and the left lane marking is a crossable lane marking, with the left lane change flag (Left_flag=1). The lane change safety assessment module outputs the flag image as shown. Figure 11 As shown in Figure 7, since there are no other vehicles in the target lane for lane changing, the lane changing safety flag3 = 1. At 36 seconds, all three module flags output 1, and the lane changing decision module's lane changing decision flag = 1. Figure 11 .8), execute lane change.

[0115] Example 3

[0116] The target lane for the test vehicle's lane change involves interfering vehicles. The scenario setup is as follows: Figure 12 As shown: In the two-lane scenario, the following vehicle is the test vehicle, traveling at a constant speed of 10 m / s along the current lane; the vehicle 190 m ahead of the test vehicle is the malfunctioning vehicle, traveling at a constant speed of 10 m / s along the current lane, slowing down until it stops due to the malfunction, with a travel distance of 60 m. The interfering vehicle in the target lane is located 160 m behind the test vehicle, traveling at a constant speed of 20 m / s along the target lane. The test road length is 600 m, and the simulation time is 30 s.

[0117] The specific parameter changes of the lane-changing decision model during the simulation are as follows: Figure 13 As shown in the image, the relative distance between the autonomous vehicle and the vehicle in front in the current lane is as follows: Figure 13 As shown in Figure 1, initially, the vehicle ahead and the test vehicle are at the same speed, and the vehicle ahead is outside the radar detection range. Due to a malfunction, the vehicle gradually decelerates and stops, and the relative distance between the vehicles gradually decreases. The test vehicle's front radar detects the vehicle ahead in approximately 9 seconds. The relative speed image between the autonomous vehicle and the vehicle ahead in the current lane is shown in Figure 1. Figure 13 As shown in Figure 2, the vehicle in front stops at 12.5 seconds, therefore the relative speed between the two vehicles remains at -10 m / s. 2 Left and right. The lane change demand judgment module calculates the collision time (TTC) and vehicle speed underrun cumulative degree (K) based on the above two parameters, as shown in the image. Figure 13 As shown in Figures 7 and 13.8, the image depicting the relative distance between the autonomous vehicle and the vehicle behind it in the target lane is as follows. Figure 13 As shown in Figure 3, the relative speed image with respect to the vehicle behind in the target lane is as follows: Figure 13 As shown in Figure 4, the image showing the relative distance to the vehicle in front in the target lane is as follows: Figure 13 As shown in Figure 5, the relative speed image with respect to the vehicle in front in the target lane is as follows: Figure 13 As shown in Figure 6.

[0118] During this process, the cumulative speed K never reached the set threshold. The collision time TTC gradually decreased as the relative distance between the two vehicles decreased. The TTC time reached the TTC threshold range at 21 seconds, triggering a lane change request. The lane change request flag bit flag2=1 was output, and the lane change request module output image as shown. Figure 13 As shown in Figure 9. At this time, the lane condition judgment module's flag output image is as follows: Figure 13 As shown in Figure 10, the lane markings are clear and valid, and the left lane marking is a crossable lane marking, with the left lane change flag (Left_flag=1). The lane change safety assessment module outputs the flag image as shown. Figure 13 As shown in Figure 11, due to the presence of interfering vehicles in the target lane, when the interfering vehicle reaches the safe area behind the test vehicle, the lane change safety judgment module determines that there is a danger and outputs the lane change safety flag3=0. After the interfering vehicle overtakes the test vehicle and enters the safe area in front of the test vehicle, the lane change safety judgment module again determines that there is a danger and outputs the lane change safety flag3=0. At 21 seconds, a lane change request is generated, at which point the lane change safety flag3=1. All three module flags output 1, and the lane change decision module's lane change flag=1. Figure 13 .12), execute lane change.

[0119] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent.

Claims

1. A lane-changing decision-making system based on logical rules, characterized in that, include: Lane condition judgment module, lane change demand judgment module, lane change safety judgment module, lane change decision module, and sensor information preprocessing module; The lane condition determination module is used to determine lane conditions; The lane change requirement determination module is used to determine lane change requirements; The lane change safety assessment module is used to assess lane change safety. The lane-changing decision module is used to make a comprehensive decision based on the judgment results of lane conditions, lane-changing needs, and lane-changing safety. The lane change demand determination module receives current lane vehicle information filtered by the sensor information preprocessing module: relative distance and relative speed. The driver's lane change demand is affected by the following factors: The driver's desired speed is quantified by the vehicle speed dissatisfaction and the cumulative vehicle speed dissatisfaction. There is a possibility of collision with the vehicle ahead; the occurrence of lane-changing intention is measured by the TTC collision time. Input the final calculated vehicle speed dissipation accumulation and TTC collision time into the Matlab Function module. When either the vehicle speed dissipation accumulation or the TTC collision time is within the threshold range, it is determined that a lane change requirement has been generated. The Matlab Function module outputs the corresponding lane change requirement flag. The formula for calculating the vehicle speed non-compliance is: ； Where K represents the vehicle speed dissipation over a period of time, and e is the gain coefficient. The current vehicle speed, For the vehicle's desired speed, This is the upper limit of vehicle speed error. The sampling time interval; The formula for calculating the cumulative percentage of vehicle speed not being fully accumulated is: ； in, The unsatisfactory cumulative rate of vehicle speed after the nth sampling time interval; When the vehicle speed is less than the cumulative threshold Km, that is: The intention to change lanes arises; The formula for calculating TTC collision time is: ； Where S is the distance between the obstacle in front and the vehicle. The speed of the obstacle relative to the vehicle; When TTC is within a certain range, that is: The intention to change lanes arises; The lane change safety judgment module receives target lane vehicle information filtered by the sensor information preprocessing module: the relative distance and relative speed of the vehicle in front detected by the front radar sensor, and the relative distance and relative speed of the vehicle behind detected by the rear radar sensor. The lane change safety judgment module plans a lane change safety zone in the target lane to judge the safety of the vehicle lane change. The safety zone is represented by the safe distance of the vehicle in front and the safe distance of the vehicle behind. The Matlab Function module predicts the safety zone in front and behind respectively and outputs the corresponding safety flag. The formula for calculating the safe distance from the vehicle ahead is: ； The formula for calculating the safe distance from vehicles behind is: ； in, This indicates the longitudinal distance between the test vehicle and the vehicle in front. This represents the longitudinal distance between the test vehicle and the vehicle behind it, where L represents the length of the test vehicle and D represents the lane-changing safety distance threshold. Indicates the longitudinal speed of the test vehicle. Indicates the longitudinal speed of the vehicle ahead. This indicates the longitudinal speed of the vehicle behind.

2. The lane-changing decision system based on logical rules according to claim 1, characterized in that, The lane condition judgment module judges lane conditions based on the analysis of lane line marker sensor data, including: the Matlab Function module outputs the corresponding left / right lane change markers based on the validity of the lane lines in the left and right lanes of the vehicle and the lane ID of the lane lines.

3. The lane-changing decision system based on logical rules according to claim 1, characterized in that, The inputs for the lane change decision module to determine whether to brake include the left / right lane change flags output by the lane condition judgment module, the lane change demand flags output by the lane change demand judgment module, the lane change safety flags output by the lane change safety judgment module, and the TTC collision time. Based on the lane change rules, the lane change decision module outputs corresponding follow, lane change, and braking commands through the inter-level state transitions of the state machine.

4. A lane-changing decision-making method based on logical rules, based on the lane-changing decision-making system based on logical rules as described in any one of claims 1-3, characterized in that, Includes the following steps: Determine lane conditions based on sensor data; Determine lane-changing needs based on sensor data; Determine lane change safety based on sensor data; Make a comprehensive decision based on the assessment results of lane conditions, lane changing needs, and lane changing safety; Three key lane-changing scenarios were constructed using Prescan, and lane-changing decision models were built in Matlab / Simulink. The lane-changing decision results for each scenario were analyzed to verify the feasibility and reliability of the lane condition judgment module, lane-changing demand judgment module, lane-changing safety judgment module, and lane-changing decision module, thus completing the lane-changing decision process for autonomous vehicles.

Images