Optimization of Elevator Usage by Image Processing for Optimal Energy Conservation

An efficient vertical transit system is a crucial feature of contemporary office buildings. Modern machine learning (ML) algorithms make it simple to optimise lift control strategies. This study proposes a revolutionary way to use Raspberry Pi camera module data for the best possible dispatching of traditional passenger lifts. It is assumed that an image processing system processes a real-time video to ascertain the quantity of people and objects utilising the lifts and waiting for a lift vehicle in the halls. These numbers are assumed to be connected to a specific uncertain probability. The efficiency of our unique lift control algorithm is derived from the need to serve a crowded floor completely, sending as many lifts as possible there and filling them to the maximum weight permitted, in addition to the probabilistic utilisation of the number of people and/or items waiting. The suggested technique introduces the idea of the effective number of persons and items to account for the uncertainty that may arise from the image processing system's imperfection. Reducing wait time, energy conservation and optimization are main the goals of this research. The proposed approach was implemented, and the simulation results showed that the passenger journey time was reduced. A three-story office building was the intended application for the prototype model. Key Words: Elevator, Raspberry Pi, Machine Learning, Optimization, Image Processing