论文部分内容阅读
为缓解交通堵塞,基于人工智能的强化学习理论,提出了不完全信息下的强化学习单点入口匝道控制方法(RLRM)。基于6个仿真实例,分别计算了平均速度、平均密度、流出交通量与旅行时间,比较了无控制、定时控制与RLRM控制的控制效果。仿真结果表明:在交通量较小的实例1中,以旅行时间为评价指标,定时控制与RLRM控制的交通阻塞缓解率分别为-6.25%、-9.38%,几乎没有控制效果;在交通量变大的实例3中,以旅行时间为评价指标,定时控制与RLRM控制的交通阻塞缓解率分别为-8.19%、3.51%,匝道控制有一定效果,RLRM控制略优于定时控制;在交通量最大的实例6中,以平均速度、平均密度、流出交通量与旅行时间为评价指标,定时控制的交通阻塞缓解率分别为8.20%、0.39%、18.97%与23.99%,RLRM控制的交通阻塞缓解率分别为18.18%、3.42%、30.65%与44.41%,RLRM控制明显优于定时控制。可见,交通量越大,RLRM控制效果越明显。
In order to alleviate the traffic jam, based on the reinforcement learning theory of artificial intelligence, an enhanced learning single point entry ramp control method (RLRM) under incomplete information is proposed. Based on the six simulation examples, the average velocity, average density, outflow of traffic and travel time were calculated, and the control effects of no control, timing control and RLRM control were compared. The simulation results show that in the case of small traffic volume 1, the travel time as the evaluation index, the traffic jam response rate of time control and RLRM control are -6.25% and 9.38%, respectively, with almost no control effect. When the traffic volume becomes larger In the example 3, taking the travel time as the evaluation index, the traffic jam remission rate of the timing control and the RLRM control is -8.19% and 3.51% respectively. The ramp control has a certain effect and the RLRM control is slightly better than the timing control. In the traffic with the largest traffic volume In Example 6, the average speed, average density, outflow of traffic and travel time are the evaluation indexes. The traffic jam remission rates under the regular control are 8.20%, 0.39%, 18.97% and 23.99% respectively. The RLRM control traffic congestion remission rates are respectively 18.18%, 3.42%, 30.65% and 44.41% respectively. The RLRM control was significantly better than the timing control. Can be seen, the greater the traffic, RLRM control effect more obvious.