/TSP_SimulatedAnnealingImproved

An implementation of Simulated Annealing algorithm for solving TSP problem. By Siddhant Kumar and Neha Muthiyan, for the course CS305-Artificial Intelligence at IIT Mandi, under the guidance of Dr. Deepak Khemani

Primary LanguageC++MIT LicenseMIT

TSP_SimulatedAnnealingImproved

This is an implementation of TSP problem in C++ using Simulated Annealing algorithm.
I have made a slight modification in the algorithm which increases the performance, It essentially increases the exploration if the algorithm is stuck at a sub-optimal solution for a long time.
This project was developed as an assignment of Artificial Intelligence (CS305) course at IIT Mandi, under the guidance of Dr. Deepak Khemani.

Usage: To run the sample test cases-
g++ -o tsp tsp.cpp
./tsp <"input_file"

To use on a general adjacency matrix-
include TSP_SA.h in your program
create an object of TSP_SA class the constructor for which is-
TSP_SA(double **adjacency_matrix, int num_of_cities);
call begin() function on the object

eg: TSP_SA TSP(edges,n);
TSP.begin();

It prints the best path at every instance