Elevator Project

Alvis Abreo

Alvis Abreo

Bengaluru, Karnataka

The students always tend to have the issue of figuring out which lift to use to reach their class floor at the peak hours. But to choose the optimal elevator could be worrisome and time consuming. But with the use of this AI/ML model that can figure out which lift to take to reach the desired floor ...learn more

Project status: Under Development

oneAPI, Artificial Intelligence, Cloud

Intel Technologies
oneAPI, DevCloud

Code Samples [1]Links [1]

Overview / Usage

This project aims to optimise elevator traffic in a university or college setting, with the goal of improving the experience for students. The problem addressed is the issue of elevators becoming overcrowded, leading to long wait times and a frustrating experience for students. By using machine learning and artificial intelligence techniques, we propose to develop a system that can predict elevator usage patterns. The proposed system has the potential to greatly improve the student experience in university or college buildings by reducing wait times and improving elevator traffic flow.

This model takes the queue quantity and your desired floor as input where queue quantity is taken from the cameras which is using AI and ML tools to find the number of people standing in the queue as head counts and then the floor number is given by the users as per the floor they need to reach,with that we have formulated a logic to find out the time every lift is going to take to reach that particular floor and the lift taking the least time to reach the desired floor is the output .`

Methodology / Approach

  • Collected data of the time taken by each lift to reach each floor in different scenarios
  • Then processed the data to find average time taken by the lifts to reach each certain floors
  • Along with the average timings of each lift we consider the queue size in front of each elevator and average capacity of a lift to calculate which elevator is the most efficient to reach a certain floor.
  • The model takes the queue quantity in front of each elevator and the destination floor as the input and provides the elevator option to take, which is calculated based on the above formulae.
  • We have used queue image processing to take the count of people in the queue.
  • Train and Test the model using intel oneAPI

Repository

https://github.com/cratonoid/Elevator_Project.git

Collaborators

3 Results

3 Results

Comments (0)