Air-conditioning operation terminal, computer readable medium and air-conditioning system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing air-conditioning apparatuses with multiple vanes lack the ability to determine and operate specific vanes for adjusting air direction and volume, as previous technologies only address single-vane systems.
Innovation Solution
An air-conditioning operation terminal that includes an image acquisition unit, object detection unit, vane selection unit, image display unit, designation acceptance unit, and air-conditioning setting unit, utilizing a learned model for vane detection and adjustment interface to allow users to select and adjust vanes in multi-vane systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a terminal device is used to operate air-conditioning apparatus with multiple vanes, then user convenience is improved, but the ability to determine and operate specific vanes is lost
Solution Approach 1:
The system segments the control interface by detecting and displaying individual vanes separately. Each vane is identified and presented as a distinct selectable element, allowing users to operate specific vanes independently rather than controlling all vanes as a single unit.
Solution Approach 2:
The terminal device acts as an intermediary between the user and the air-conditioning apparatus. It captures images, detects vanes using machine learning models, and translates user selections into specific control commands for targeted vanes, bridging the gap between simple terminal operation and complex multi-vane control.
2Measurement precision
If machine learning is used for vane detection, then detection accuracy is improved, but processing time is increased
Solution Approach 1:
The machine learning model is trained in advance on a large dataset of air-conditioning unit images with annotated vanes. This preliminary training phase enables the model to quickly and accurately detect vanes during actual operation without requiring complex real-time processing, as the feature extraction patterns are already learned.
Solution Approach 2:
The system uses a pre-trained machine learning model that has learned vane patterns from training images. Instead of performing complex real-time learning, the system applies the learned model to detect vanes in captured images, copying the detection capability from trained data to operational data efficiently.
Data Source
AI summary
An object detection unit (212) uses a learned model to detect a plurality of vanes in a captured image in which an air-conditioning indoor unit is captured. A vane selection unit (214) selects a target vane from the plurality of vanes in the captured image. An image display unit (215) displays a superimposed image in which a target identification mark and an adjustment interface are superimposed. A designation acceptance unit (216) accepts adjustment details designated by operating the adjustment interface. An air-conditioning setting unit (217) sets the accepted adjustment details in the air-conditioning indoor unit.