Waste collection vehicle navigation in modern cities

Abstract

It is evident that over the last years the usage of robotics and automated machinery in our society has been rapidly increasing. Specifically, the usage of automated vehicles in industrial and also everyday life has benefited all of humanity greatly. One area that has not been fully explored is the usage of automated robotic vehicles in waste management within cities. It is clear that the rate of urban waste production is constantly increasing as a result of the Earth's rapid population growth and modern lifestyle. People consume more and more, and as a result, they produce more and more waste. In most countries, these waste materials are being thrown into bins in the streets of the cities. Opposed to the traditional waste collection model where a waste collection truck, passes from all smart bins and picks up the waste materials in a pr- defined route, we propose an algorithm to dynamically generate these routes in real time, based from the data received from the installed smart-bin devices. The vehicles, however, are only as efficient as the algorithms that govern them. The majority of the waste collection algorithms found in literature, however, are statically designed and cannot handle multi-robot situations and utilize data received in real time. The proposed algorithm of this paper attempts to give a solution to this issue; utilizing more than one waste collection vehicles simultaneously to allocate waste pick up tasks and tailor the navigation path of each vehicle based on its characteristics, like its type and its current location within the environment so as to minimize the pick up timing. We evaluated the proposed methodology in a synthetic realistic environment and and demonstrated that the algorithm is capable of finding an improved solution within a realistic time frame.