关键词: 匝道, 元胞自动机, 仿真, 车联网中图分类号:U491 文章标识码:A 文章编号:1007-3221(2024)04-0140-07 doi:10.12005/orms.2024.0124

Abstract: With the rapid development of the social economy, the optimized driving mode of vehicle X interconnection has improved traffic efficiency, maintained traffic safety, and become a future trend. With the introduction of the Internet of Vehicles (IoV), traffic flow simulation under the IoV has become a hot topic. As one of the bottleneck scenarios, the ramp is a classic scenario for studying the solution of traffic congestion and safety problems. However, few scholars choose to study it after the emergence of the Internet of vehicles. This paper takes the ramp as the research object and simulates the vehicle running state under the IoV environment by improving some rules.
A cellular automaton model of the upper ramp with the main road as a double lane, considering the safe distance and the interconnection of the IoV, is established, and the corresponding simulation experiments are carried out. In view of the on-ramp scenario where the main road is two-lane, the cooperative lane change rule is proposed. A virtual monitoring point is set up between the end cell of ramp section and the beginning cell of acceleration lane. The forced lane change rule is also improved. For the HVs, the HV on the acceleration lane changes lanes after independent judgment when it joins the main road. If the lane change is not possible, to continue until the lane changes is successful. If you reach the end of the acceleration lane and still do not change the lane successfully, wait for a suitable lane change opportunity at the end to change the lane. For the CAVs, the CAV on the ramp finds the location of the vehicle on the slow lane at the next time through pre-judgement, and finds the farthest position within its accessible speed range. At this time, adjust the speed of the vehicle down time, accurately reach the position and change lanes. If the appropriate point is not found, it is judged and predicted until the end of the acceleration lane is reached and the rules for changing lanes are formulated.
In order to verify the traffic efficiency and safety level under the IoV, a cellular automaton simulation is conducted for the vehicle on ramp process under the high-level IoV, simulating the traffic efficiency and safety situation under the mixed arrangement of CAVs and HVs. When the traffic density is large, the index values of different vehicle arrangements are significantly different.
In terms of traffic efficiency, the impact of different vehicle arrangements on traffic flow is relatively small at low densities, but significant at medium or high densities. When the density is medium or high, if the penetration rate of CAVs is high in the arrangement, the traffic efficiency is also high. The introduction of CAVs has a promoting effect on the growth of traffic efficiency, while the high penetration of HVs in the arrangement will inhibit the growth of traffic efficiency.
In terms of traffic safety, in the case of medium density, the possibility of collision between a CAV and a HV is higher but the collision damage is small. HV pure flow will be safe because of low speed. The rest have low collision probability but high damage. In the case of high density, the possibility of collision between three CAVs and three HVs is greater, and the degree of damage is large, which is more dangerous. In the case of high density and dense vehicles, CAVs have a significant positive effect on traffic safety. When the traffic flow is the pure flow of CAVs, the traffic situation is not only efficient, but also safe. In the case of mixed traffic, when the lane change of vehicles can be selected to arrange, the long queue of CAVs should be formed as far as possible to reduce the aggregation of HVs.

Key words: ramp, cellular automata, simulation, Internet of vehicles