| The introduction of autonomous vehicles will greatly change the current traffic system and have an impact on road traffic flow.In a mixed traffic flow composed of autonomous and manual drivers,the difference in the degree of trust in autonomous driving by manual drivers may make them exhibit different characteristics when following.Based on these two points,the research objectives of this thesis are proposed as follows:1)To study the influence of trust degree on the longitudinal characteristics of the driver in following autopilot.2)Establish a mixed traffic flow following model that can reflect the heterogeneity of drivers’trust degree.3)Based on the established mixed traffic flow following model,the influence of autonomous driving permeability and driver trust on road traffic flow is discussed.In this thesis,a virtual driving experiment platform is built based on unity3d software.Drivers with different degrees of trust are recruited for car following experiments.The characteristics of car following automatic driving and manual driving are studied,and the approximate quantitative relationship between the distance of car following automatic driving and the distance of car following manual driving is obtained.Then,according to the characteristics of automatic driving and manual driving vehicles,safe distance model and full speed difference model are used to establish car following rules respectively.Combined with the results of virtual driving experiment,the driver trust degree parameter u_m is introduced into the model,and a mixed traffic flow car following model considering the heterogeneity of driver trust degree is proposed.Finally,based on the continuous cellular automata model,the single lane micro model and intersection micro model are established in the netlogo simulation software,and the simulation experiments are carried out to explore the influence of automatic driving penetration and driver trust on the mixed traffic flow.The results show that:1)the degree of trust has a significant impact on the car following distance of drivers in mixed traffic flow.The distance of drivers with low trust,medium trust and high trust in car following automatic driving is 123.0%,94.5%and 79.2%of that in car following manual driving,respectively;Trust has no significant effect on the parking distance of drivers in mixed traffic flow.2)In a single-lane mixed traffic flow with a certain density,increasing the permeability of autonomous driving can increase the traffic flow and average speed by about 3.5 times on average.In the single-lane mixed traffic flow with a certain density,when the autonomous driving penetration rate is low,medium and high respectively,the improvement of driver trust can increase the traffic flow and average speed by 19%,24%and 11%on average.3)In the mixed traffic flow at intersections,when the ratio of drivers with different levels of trust is 1:1:1,increasing the permeability of autopilot can reduce the average delay by 1.58s at most;When the autonomous driving penetration rate is 0.5,improving the trust of drivers can reduce the average delay at the intersection by 1.2s. |
