3D Object Detection for Safe AMR Elevator Entry Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous Mobile Robots (AMRs) face challenges in determining suitable times to enter elevators due to potential obstacles like wheelchairs or strollers, which may hinder smooth exit of larger objects.
Innovation Solution
An object detection device and method using image-extraction and depth imaging to identify objects, calculate their depth values, and determine their status, allowing the AMR to decide when it is safe to enter a space by considering object presence and status.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the AMR waits directly in front of the elevator door or directly enters the elevator, then the AMR can quickly access the elevator, but larger objects (wheelchairs, strollers) cannot exit the elevator smoothly
Solution Approach 1:
The system performs preliminary detection of objects inside the elevator before the AMR enters. By using image extraction and depth imaging to identify wheelchairs, strollers, or other large objects in advance, the system can determine the appropriate waiting time or entry timing to ensure smooth exit of larger objects while maintaining efficient access for the AMR
2Loss of time
If the AMR enters the elevator immediately, then the AMR minimizes waiting time, but the system cannot determine if it is safe to enter
Solution Approach 1:
The system replaces manual safety checking with automated image extraction and depth imaging technology. The processor analyzes captured images to detect objects and calculate their positions, automatically determining whether it is safe for the AMR to enter the elevator without requiring manual intervention or extending waiting time
3Device complexity
If the system uses simple presence detection, then the detection system is simple, but it cannot determine object status or appropriate entry timing
Solution Approach 1:
The system transitions from two-dimensional image detection to three-dimensional spatial understanding by incorporating depth imaging. This allows the processor to calculate actual distances and positions of objects within the elevator space, providing comprehensive object status information including location, size estimation, and potential movement paths, enabling accurate entry timing determination
Data Source
AI summary
An object detection device is provided in the invention. The object detection device includes an image-extraction device and a processor. The image-extraction device extracts a plurality of images and a plurality of depth images in a specific space, wherein each image corresponds to one of the depth images. The processor may detect objects in each image using an object detection model to obtain object information. The processor may calculate the depth value of each object in each depth image. The processor may input the depth value of each object into an object status model to determine the object status of each object. The processor may determine whether to enter the specific space based on the object information and the object status of each object. Therefore, the object detection device provided in the invention will be able to determine a suitable time to enter the specific space.