UAV Delivery Image Capture for Verifiable Package Drop-Off
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In unattended package delivery by unmanned vehicles, capturing an appropriate image to prove delivery completion is challenging, especially when the package is placed at any position in a large site, as the deliverer cannot photograph the package.
Innovation Solution
The unmanned vehicle includes a camera and control units to image a range within its view, controlling its position and orientation to ensure the package and its peripheral region are captured, and saves an image that proves delivery completion, optionally with additional features as marks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If unattended delivery is performed by unmanned vehicle without deliverer, then delivery efficiency is improved, but ability to capture delivery completion image deteriorates
Solution Approach 1:
The unmanned vehicle performs delivery and image capture autonomously without human deliverer intervention. The vehicle itself captures images of the delivered package and surrounding environment, and automatically determines delivery completion based on image analysis, eliminating the need for a human deliverer to photograph the package.
Solution Approach 2:
The patent replaces the mechanical action of a human deliverer photographing the package with an automated image capture system. The unmanned vehicle uses its onboard camera to capture images and automated image processing algorithms to determine delivery completion, substituting human mechanical actions with automated systems.
2Adaptability or versatility
If package is placed at any position in large site, then delivery flexibility is improved, but ability to capture recognizable image deteriorates
Solution Approach 1:
The patent transitions from capturing only the package in a two-dimensional view to capturing the package within its three-dimensional surrounding environment. By photographing not just the package but also the surrounding area including buildings, roads, and other landmarks, the system creates contextual spatial information that enables precise position identification even when the package is placed anywhere in a large site.
Solution Approach 2:
The surrounding environment (buildings, roads, landmarks) acts as an intermediary reference system. Instead of relying solely on the package itself for position identification, the system uses the surrounding environment as reference objects to determine and verify the package's location, enabling accurate position identification in large areas.
3Reliability
If image processing is performed to detect features, then delivery proof accuracy is improved, but processing load and power consumption increase
Solution Approach 1:
The patent applies partial image processing by focusing computational resources on detecting specific features (buildings, roads, landmarks) that are most relevant for position verification, rather than processing the entire image in detail. This selective approach maintains delivery proof accuracy while reducing overall processing load and power consumption.
Data Source
AI summary
In a delivery system S including an UAV 1 and a management server 2, the UAV 1 controls at least a position and/or an orientation of the UAV 1 so that a package placed in a release location and a peripheral region of the package fall within an angle of view of a camera. And then, the UAV 1 saves, as an image that proves delivery completion of the package, an image of the peripheral region including the package captured by the camera in a storage unit 15 of the UAV 1 or a storage unit 22 of the management server 2.