Artificial intelligence project

Description In this project you will write a program to plan a path from the source to the goal in a given environment represented by a grid. Your input would be a matrix text file ([login to view URL]) like this: WPPWWW WWPWPS WWPWPW PPPPPW GWPWWW W is a wall, P is a path, S is the source and G is the goal. Note that the size of the matrix can be any dimensions larger than 4, and the number of rows and columns are not always the same. For the input above the output matrix is as follows; where stars indicate the path solution from S to G: WPPWWW WWPW*S WWPW*W *****W GWPWWW For the above example the path solution is {S, (1,4), (2,4), (3,4), (3,3), (3,2), (3,1), (3,0), G} and its cost is 8. Moreover, in some input files, there will be different types of objects denoted as any capital letter other than the four above for example: WPPWWW WWPWPS WWPWBA PPPCPW GWPWWW A, B and C are images, where the name of the image files are stored in another input file ([login to view URL]) with the following structure, where the letters: A [login to view URL] B [login to view URL] C [login to view URL] You will use these images (the image files will be given to you as input) and classify them as either obstacles or paths depending on the content of the image. To find the path solution you will use two versions of A* algorithm implementing the straight line distance (h1) and the Manhattan distance (h2) heuristics respectively. Also to classify the images you will use the pretrained deep learning model called VGG16 to identify what the image represents, which can be one of 1000 classes (labels) listed here: [login to view URL] For simplicity assume the first 398 from (0-397) are obstacles, the others are not. You can change this assumption if you want to analyze the labels further. The programming language would be python, in section 4 supplements, you can find resources further resources for help. Samples are available with this description, and test files will be sent a week before the due date.