00:04
Hello my namemiship and my partner my group partner names th pro we developed this project with named is adapted traffic signal controlling reinforcement landending basically this project based on the traffic control traffic light controlling on the basis of on the basis of traffic。For example if we have a if we have a bunch of traffic on the left side and there is no traffic on the right side so basically we are using we are using the agent this particular base agent project like a for example let me show you code first what we're using and I will explain the。Other so in this code we are using in the code are using cancerlo cares dum by DQ and so laborbr so librarybrities used for the simulation purpose anddiord we have a To Classes one name is a duanna and the other Bo sum interactction SIM interactction is used for the simulation purpose which I show you later in in the code。
01:21
Let me explain you something。Let me when you go a little bit。So basically our project is based on。Traffic control。So as you can see this is that project simulation we have a lots of we have lots of vehicle on four different routes we got a four roads right side left right up poor and down so if you see。And if we have a bunch of traffic on the left and there are minimum on the right so our project works on。A number of vehicle it will it will post countunt the number of vehicle on the left side left side of the road and right side of the road it will have a lots of card on the left side of the code left side on the road it will automatically stop the right right side of the right right side of the right side of the road by using a re reinforcement learning stop。
02:22
Right side road vehiclehi with the red light。And it will automatically green the the left side road car Suan card。嗯。So that left side road people can start run their car or anything so what is the scope of this project and five in this project is useful if you use this system we can save a lot of time normally normally waiting time is in traffic around ten to five with five but。
03:01
If you use this project if you use this system in normal roads we can save a lot of time and lots of fuel。So this is efficient efficient way of saving the time saving the time in the field so。How wealc could it be。Traffic low using the reinforcement learning physical recouned the re countuned the number vehicles number of vehicle road sites and the velocity velocity speed of the vehicle and the position of the vehicle so become the first become the number of vehicle and speed and position and then we summarize all these all these number and we use a reward based system so if you have a number a bunch of car on the left side we counted all calculate all the calculations and give them our re report rewards of zero one base reward if if you have a bunch of traffic then wewarded it to one and if we have a zero traffic then wewarded at the zero。
04:07
So then a of the code at design of the code we summarize if we have a number of if we have a maximum number of the world on the left side then we release the left side traffic and if we have a minimum number of reward on the right side then we stop the right sidero so basically this is a whole idea of project which are very simple but we are using all the。We're using all the cover all the topic which we are all learning your plan and your course so we're using a different type of agents reinforcement learning different different Python libraries we don't have any data side in it because we'we didn't need any data set here this is a simple code。So that's all from。
我来说两句