1 00:00:00,000 --> 00:00:06,120 One of our approach to solve the maze for the robot is to have the maze on the ground. 2 00:00:06,330 --> 00:00:07,320 This is the ground. 3 00:00:07,650 --> 00:00:11,550 We have a maze here having certain things. 4 00:00:11,860 --> 00:00:15,210 A robot is at this point starting this thing. 5 00:00:15,570 --> 00:00:18,000 And we have a very confusing means. 6 00:00:18,450 --> 00:00:20,280 The robot has to develop. 7 00:00:21,030 --> 00:00:30,300 Apart from these two things, I think this is not visible at this point to end point and this way. 8 00:00:30,630 --> 00:00:33,720 So what I am trying to say is we have a maze. 9 00:00:34,470 --> 00:00:35,040 Maze. 10 00:00:35,490 --> 00:00:43,230 We have a robot here and we are going to solve it using open CV to solve it to open CV when we don't 11 00:00:43,230 --> 00:00:44,460 have a map of the maze. 12 00:00:44,760 --> 00:00:47,940 We have to look at the maze to solve. 13 00:00:48,150 --> 00:00:49,740 The approach is going to be simple. 14 00:00:50,340 --> 00:00:53,910 A camera is going to be installed at this much height. 15 00:00:54,480 --> 00:00:58,980 From this point to this point, there is going to be a certain height in the z-axis. 16 00:00:59,340 --> 00:00:59,700 Okay. 17 00:00:59,700 --> 00:01:08,580 The camera is going to be looking from up from upper portion, from camera. 18 00:01:08,580 --> 00:01:11,100 It's going to be looking at from top view. 19 00:01:11,490 --> 00:01:22,080 And it is going to look at all of the maze and the robot where the robot is, what is going to be the 20 00:01:22,080 --> 00:01:23,250 structure of the maze. 21 00:01:23,490 --> 00:01:28,320 And basically it is going to obtain the map of the maze. 22 00:01:28,320 --> 00:01:34,440 So it is going then it is going to be able to solve the maze and ten commands to the robot. 23 00:01:34,860 --> 00:01:36,990 What part should it follow? 24 00:01:37,440 --> 00:01:41,640 Once it's done, it is going to then give a feedback. 25 00:01:41,640 --> 00:01:42,900 That job is done. 26 00:01:43,200 --> 00:01:47,370 In our case, currently, we have to focus on two things. 27 00:01:47,400 --> 00:01:51,930 We need to create an object, a square object, a camera looking like object. 28 00:01:51,930 --> 00:01:58,980 We can obviously have an SDL mesh from a Fusion 360 card software representing a camera, but I will 29 00:01:58,980 --> 00:02:06,540 not go into that much to trouble you if you are not if you are not good with 3D modeling. 30 00:02:06,540 --> 00:02:14,940 So basic box is going to be created and in that box we are going to add one thing which is eggs, power 31 00:02:14,940 --> 00:02:17,070 and simulation, that is plugins. 32 00:02:17,430 --> 00:02:24,990 So plug in of camera is going to be installed in this box containing some lines of code, which is going 33 00:02:24,990 --> 00:02:27,960 to give us the functionality of a camera. 34 00:02:28,290 --> 00:02:34,650 The how I am going to do that, I am going to write this code into the box first I will create a box, 35 00:02:34,650 --> 00:02:38,430 then I will write this code in the link of the box. 36 00:02:40,320 --> 00:02:44,910 What this code is saying we need to have a link to which it should reference. 37 00:02:45,240 --> 00:02:48,450 In our case, it is going to be a box or something like that. 38 00:02:48,750 --> 00:02:56,910 Then we are going to define a camera update raid, meaning to DFP s and horizontal view. 39 00:02:56,920 --> 00:02:59,580 What is going to be the horizontal view of the image? 40 00:02:59,580 --> 00:03:07,410 This is the length of the horizontal view and then length and the height and weight of the image. 41 00:03:07,410 --> 00:03:10,170 800 by 800 is going to be the image. 42 00:03:10,170 --> 00:03:14,430 We are going to be increasing it to 1200 forward is going to be RGV. 43 00:03:14,430 --> 00:03:22,520 So basically we are defining properties of one single image and as it is going to be a 30 FP image produce 44 00:03:22,530 --> 00:03:30,240 that it is going to produce a video and we can set a lot of things in the simulation about the properties 45 00:03:30,240 --> 00:03:33,870 of the camera as it is a simulation and we are going to do that. 46 00:03:34,500 --> 00:03:38,010 We need this video to be saved for the computer vision task. 47 00:03:38,370 --> 00:03:44,790 We are not going to use the approach of applying computer vision algorithm directly on a real time system. 48 00:03:45,150 --> 00:03:52,200 We are going to extract the video out of it, recording it by recording it, and then we will provide 49 00:03:52,200 --> 00:03:58,290 it to we can provision DB, which is going to perform all of the stuff related to the path planning 50 00:03:58,290 --> 00:04:04,470 and edge detection and motion planning and waypoints finding and stuff like that. 51 00:04:04,470 --> 00:04:10,530 Then it will return us the next move the robot should take and we will implement that. 52 00:04:10,530 --> 00:04:12,960 So this is going to be the basic framework. 53 00:04:13,200 --> 00:04:18,000 Let's take a look how we are going to make the camera work and save the video out of it.