Elevator Passenger Detection for Social-Distancing Dispatch Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Elevator systems face inefficiencies due to inaccurate detection of waiting passengers, leading to unnecessary stops, power consumption, and potential spread of communicable health conditions, as conventional motion detectors provide limited accuracy and cannot distinguish between individuals and objects.
Innovation Solution
Implementing machine learning and deep learning algorithms to process images for accurate identification of waiting passengers, allowing the elevator system to adjust dispatch and capacity based on actual passenger presence, thereby optimizing operations and promoting social distancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional motion detectors are used to detect waiting passengers, then the system can identify passenger presence, but the detection accuracy is limited and cannot distinguish between individuals and objects
Solution Approach 1:
The patent replaces conventional motion detectors with image capture devices (cameras) combined with machine learning algorithms. This substitution enables the system to visually identify and count individual passengers by analyzing image data, thereby achieving accurate distinction between people and objects while maintaining reasonable system complexity through software-based processing.
2Productivity
If elevators stop at every floor with call requests, then all potential passengers can be served, but unnecessary stops increase power consumption and travel time
Solution Approach 1:
The patent implements preliminary action by capturing images and identifying passengers before the elevator arrives at each floor. The system pre-determines which floors have waiting passengers and plans the stopping sequence in advance, allowing the elevator to skip floors with no passengers and reduce unnecessary stops, thereby lowering power consumption and improving service efficiency.
3Loss of time
If multiple elevator cars are dispatched to serve waiting passengers, then passenger waiting time is reduced, but system coordination complexity increases
Solution Approach 1:
The patent employs feedback mechanisms where the central controller continuously monitors passenger detection results, elevator car positions, and floor call requests. Based on this real-time feedback, the system dynamically adjusts the dispatch of multiple elevator cars, optimizing their allocation to minimize passenger waiting time while managing coordination complexity through centralized control and adaptive decision-making.
Data Source
AI summary
Provided are embodiments for identifying waiting passengers outside an elevator using deep learning and machine learning and automatically supporting social distancing for elevator passengers. Embodiments include receiving a call request from an elevator floor; capturing an image of the elevator floor registering the call request responsive to receiving the call request; and determining, from the image, a presence of one or more waiting passengers on the elevator floor registering the call request. Embodiments also include dispatching an elevator car to the elevator floor registering the call request; and controlling the elevator car based at least in part on determining the presence of the one or more waiting passengers.