Signal and Data Processing

Abstract

In today's world, urban traffic networks are facing many challenges. And the use of artificial intelligence and the improvement of routing algorithms and its localization in the urban sector can be efficient in the organization of urban management. In recent years, due to urban growth and development, the need for safe and optimal transportation has increased in the field of urban logistics issues. One of these issues is the transportation of money with armored vehicles of banks, which is a risky business and requires careful planning and supervision. Therefore, in addition to being on the shortest route, the optimal time and maintaining the security of armored vehicles of banks in transferring money is very important in the discussion of urban development. The main challenges of cash machines are security, timing, traffic and optimal route selection. In this regard, the concept of intelligent routing has been proposed as an efficient solution. At the heart of every intelligent routing system is a routing algorithm that performs routing from the origin to different destinations according to the determined strategy and setting priorities. A good route is usually the one that costs the least. There are different routing algorithms, here we will examine the concept of ant colony algorithm for a better understanding of the article. Ant colony optimization, abbreviated as ACO, proposed by Marco Dorigo in 1992, is one of the most prominent algorithms for collective intelligence methods. This algorithm is inspired by the collective behavior of ants. Ants work together to find the shortest route between the nest and food sources so that they can transport the food to the nest in the shortest time. Ants leave a trail of chemical pheromone when they move along the path, of course, this substance soon evaporates, but in the short term, it remains on the ground as an ant trail. There is a simple basic behavior in ants: when choosing between two paths, they are likely to choose the path that has more pheromone, or in other words more ants have already passed it. It should be noted that this simple arrangement leads to finding the shortest path. According to the research conducted in smart routing in urban traffic, despite its advantages and applications, there are challenges such as not paying attention to critical points such as route security, etc., which is one of the factors of increasing the time. Research shows that the focus is only on finding the shortest path without considering the critical paths, while one of the most important goals in finding the shortest path is to reach the optimal time and maintain security. In this research, a collective intelligence approach called the ant algorithm is used to find the most optimal route by considering more exploratory information including high traffic, low security and dangerous driving (dangerous road) as critical routes that increase time and decrease security. Therefore, the most important aspect of innovation and the most recent of this research is the use of more heuristic information in the optimization of the ant algorithm in solving the traveling salesman problem (TSP) in finding the optimal Hamiltonian path. The main purpose of this practical article is to provide a method for intelligent routing of cash machines using the ant colony optimization algorithm using the data of Shiraz urban traffic map located in Fars province, including eight banks and the routes leading to them. In this implementation, using the Folium web application, the data that includes coordinates and geographical positions is depicted, and the distance between two points is calculated using the Haversine method using longitude and latitude based on the laws of spherical geometry. to be The proposed algorithm based on the ant colony optimization algorithm with 8 nodes and 30 artificial ants in the number of iterations, 10, 50, 100, and 200 and dynamic adjustment of parameters alpha 5. 0.0, 0.3, 0.1, beta 0.5, 0.3, 1.5 and evaporation rate 0.5, 0.3, and 0.1 and considering more heuristic information such as heavy traffic, low security, and unsafe driving as a critical path by setting a penalty for the probability of being on these paths solves the TSP problem. Routing is done and the algorithm finds the optimal Hamilton circuit. The proposed method can choose a route for cash machines that avoid crime-prone areas, heavy traffic, and road hazards, and reach the destination faster and more efficiently by saving time. Considering the importance of smart routing, this article can help to improve and be productive in safe transportation and be used as an effective solution in this field.