CN-121973776-A - Intelligent vehicle ramp afflux decision method considering advanced afflux of rear vehicles to main line and vehicles behind main line

CN121973776ACN 121973776 ACN121973776 ACN 121973776ACN-121973776-A

Abstract

The invention belongs to the technical field of intelligent automobiles, and particularly relates to an intelligent vehicle ramp entry decision method considering that a rear vehicle enters a main line in advance and vehicles behind the main line, which comprises the steps of acquiring the road level position, the speed and the longitudinal acceleration of a self-vehicle, and judging that the self-vehicle enters an active entry decision stage if the three meet set conditions; the method comprises the steps of carrying out panoramic scanning on an accelerating lane behind the vehicle, screening potential cooperative vehicles according to a set mode, further monitoring behavior characteristics of the potential cooperative vehicles, judging that the potential cooperative vehicles have advanced import behaviors when set conditions are met, immediately starting safety evaluation, and if the potential cooperative vehicles are judged to be safe advanced import and the collision time of the vehicle and the potential cooperative vehicles is greater than or equal to 0.8s, completing import by the vehicle according to an acceleration curve greater than that of the potential cooperative vehicles, otherwise, executing a conservative import strategy. The method can improve the success rate, safety and efficiency of intelligent vehicle import.

Inventors

WANG CHANG
HAN YAXI
LI MINGJUN
WANG XIAO

Assignees

长安大学

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (8)

1. An intelligent vehicle ramp import decision method considering a rear vehicle import main line in advance and a vehicle behind the main line is characterized by comprising a road grade position, a vehicle speed and a longitudinal acceleration of a self vehicle, judging that the self vehicle enters an active import decision stage if the road grade position, the vehicle speed and the longitudinal acceleration of the self vehicle meet set conditions, simultaneously starting a millimeter wave radar and a vision sensor of the self vehicle, carrying out panoramic scanning on an acceleration lane behind the self vehicle, screening potential cooperative vehicles according to a set mode, further monitoring behavior characteristics of the potential cooperative vehicles, judging that the potential cooperative vehicles have import-in advance behaviors, immediately starting safety evaluation, and carrying out a conservative import strategy if the potential cooperative vehicles are judged to be the safe import-in advance, and the collision time of the self vehicle and the potential cooperative vehicles is more than or equal to 0.8s, finishing the import-in with an acceleration curve which is more than that of the potential cooperative vehicles; The potential cooperative vehicles are nearest neighbor vehicles which are in the same lane behind the vehicle and have motion tracks related to the vehicle.
2. The intelligent vehicle ramp merge decision method of claim 1, wherein the on-board system determines that the vehicle has entered an active merge decision stage when the vehicle is simultaneously in an acceleration lane at a road level position, at a vehicle speed greater than 40km/h, and at a longitudinal acceleration greater than 0m/s 2 .
3. The intelligent vehicle ramp afflux decision method of claim 1, wherein the manner of screening potential cooperative vehicles is as follows: The millimeter wave radar and the vision sensor at the left rear side and the right rear side of the vehicle carry out panoramic scanning on an acceleration lane within the range of 50-200 meters behind the vehicle, and the vehicle is locked at the rear of the vehicle, in the same lane and at the distance d relative from the vehicle, wherein the distance d relative is more than or equal to 20: 20m and is less than or equal to 100: 100 m.
4. The intelligent vehicle ramp junction decision method according to claim 1, wherein the potential cooperative vehicles are continuously monitored at a frequency of 30-100 hz, and the vehicle-mounted system determines that the potential cooperative vehicles have advanced junction-in main line behavior when all of the following conditions are satisfied at the same time: the lateral acceleration ay BV > 0.3m/s 2 of the potential co-vehicle, while the head has crossed the main line target lane centerline.
5. The intelligent vehicle ramp afflux decision method of claim 1, wherein the security assessment method is as follows: The vehicle-mounted system utilizes a side backward millimeter wave radar to track a rear vehicle MV on a main line target lane, calculates collision time TTC between a potential cooperative vehicle and the MV, and carries out safety assessment according to the collision time TTC; the TTC calculation formula is: Wherein d relative , MV (t) is the longitudinal relative distance between the potential cooperative vehicle and the MV at time t, v relative , and MV (t) is the longitudinal relative speed of the two vehicles at time t.
6. The intelligent vehicle ramp merge decision method according to claim 5, wherein the potential cooperative vehicle is considered to be a safe early merge if the collision time TTC (t) > 3s is always satisfied by the potential cooperative vehicle and the rear vehicle MV on the main target lane during the process of starting the potential cooperative vehicle to move transversely from the vehicle to the process that the vehicle head crosses the center line of the main target lane, and otherwise, the potential cooperative vehicle ignores the behavior and executes a conservative merge strategy.
7. The intelligent vehicle ramp converging decision-making method according to claim 1, wherein the vehicle is capable of completing the converging process by using an acceleration curve larger than that of the potential cooperative vehicle, and the vehicle-mounted system is capable of acquiring a longitudinal acceleration sequence in a period from the start of the transverse movement of the potential cooperative vehicle to the crossing of the central line of the main line target lane by the vehicle head And calculates the average longitudinal acceleration during the import process The track planner of the own vehicle generates the same import path according to the motion track of the potential cooperative vehicle and the own vehicle is larger than the potential cooperative vehicle And accelerating the acceleration of the main line target lane by taking the entry path as a line.
8. The intelligent vehicle ramp afflux decision method of claim 7, wherein the average longitudinal acceleration The calculation formula of (2) is as follows: where t 0 is the time when the potential cooperative vehicle starts to move laterally, and t 1 is the time when the potential cooperative vehicle head crosses the main line target lane centerline.

Description

Intelligent vehicle ramp afflux decision method considering advanced afflux of rear vehicles to main line and vehicles behind main line Technical Field The invention belongs to the technical field of intelligent automobiles, in particular to an intelligent automobile ramp entry decision method considering a main line of an advanced entering of a rear automobile and a vehicle behind the main line, and particularly relates to a method for triggering or optimizing an automatic vehicle entry decision by identifying, evaluating and utilizing the preemptive entering behavior of the rear following automobile. Background In a highway ramp merge scenario, an intelligent vehicle needs to recognize an pluggable gap of a main line target lane (also called a main line/main line lane/target lane) and merge safely. The conventional import decision algorithm is mainly based on the prediction of the motion state of the target lane vehicles (i.e., the front vehicles and the rear vehicles of the main lane), and judges the import opportunity by calculating the indexes such as collision time (TTC). For example, the application number is 202010075592.9, and the Chinese patent application entitled "risk decision device and method for autonomous entrance of a vehicle into a main road in an acceleration lane" discloses calculating a real-time entrance risk level according to the remaining length of a ramp vehicle from the end of the acceleration lane, the real-time speed of the ramp vehicle, the speed of an own vehicle behind the main road, the distance between the own vehicle and the rear vehicle, and the angle between the own vehicle and the rear vehicle, and deciding the entrance mode of the vehicle according to the real-time entrance risk level. However, the overall convergence algorithm of the intelligent vehicle is more conservative, and tends to be more prone to conservative convergence decisions in some scenes with lower convergence risk, which results in a certain influence on the overall convergence success rate. For example, when traffic flow is large, the target lane may not have a sufficiently large safety gap for a long time, resulting in that the own vehicle is forced to slow down or even stop waiting at the end of the accelerating lane, which seriously affects traffic efficiency and safety. In addition, conventional algorithms treat other participants in the traffic environment as independent, or even contradictory individuals, ignoring possible interactions between vehicles. For example, when a smart car and other vehicles come in together in an acceleration lane, a scene of early entry sometimes occurs for other vehicles behind the smart car. At this time, since the rear vehicle has completed the remittance process with the risk assessment of the vehicle behind the main line, and the remittance is performed. On the premise that the decision process of the rear vehicle is confirmed to have no obvious conflict, the advanced import behavior of the rear vehicle actually forms physical blocking or psychological deterrence for the rear vehicle of the main line lane, so that a short but effective import window is created for the own vehicle, the intelligent vehicle can be used for assisting the intelligent vehicle to import into the main line earlier, and the phenomenon of import failure is avoided. The prior art can not identify and utilize the precious 'non-active cooperative signal', and the autonomous import success rate of the intelligent vehicle can not be further improved. In view of this, the present invention has been made. Disclosure of Invention The object of the present invention is to overcome the drawbacks of the prior art described above by proposing an intelligent vehicle ramp entry decision method that considers the advanced entry of a following vehicle into a main line and a vehicle behind the main line, which method enables to perceive and understand the "exemplary" preemptive entry behaviour of a following vehicle, the behavior is used as a high-value forward cooperative signal, and the incoming decision of the own vehicle is actively and safely triggered or optimized based on the signal, so that the success rate, the safety and the efficiency of incoming are improved. In order to achieve the above purpose, the present invention adopts the following technical scheme: The invention provides an intelligent vehicle ramp entry decision method considering a post-vehicle early entry main line and a vehicle behind the main line, which comprises the steps of acquiring a road level position, a vehicle speed and longitudinal acceleration of a self-vehicle, judging that the self-vehicle enters an active entry decision stage if the road level position, the vehicle speed and the longitudinal acceleration meet set conditions, starting a millimeter wave radar and a vision sensor of the self-vehicle at the same time, carrying out panoramic scanning on an acceleration lane behind the self-vehicle, screening potential