AI Robot Object Recognition and Location Mapping via Voice Command
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional robot cleaners are unable to recognize and locate small objects during cleaning, limiting their functionality.
Innovation Solution
An artificial intelligence robot equipped with image recognition capabilities that can understand speech commands and transmit the position of recognized objects to a user's mobile terminal, allowing for the creation of a cleaning map to facilitate object location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional robot cleaners use simple object recognition through camera, then they can recognize relatively large objects, but they cannot recognize and locate small objects such as rings
Solution Approach 1:
The patent segments the cleaning area into a structured cleaning map with grid cells, and further segments object detection into multiple stages: initial object detection, small object detection through image stitching, and precise location marking. This segmentation allows the system to handle different object sizes effectively by applying different detection strategies to different spatial resolutions.
Solution Approach 2:
The patent transitions from two-dimensional camera images to a three-dimensional cleaning map representation with precise coordinate information. By mapping detected objects to specific grid cells and coordinates on the cleaning map, the system enhances location precision and enables accurate object positioning that goes beyond simple image recognition.
2Adaptability or versatility
If robot cleaners perform only cleaning function, then the device complexity is low, but the functionality is limited
Solution Approach 1:
The patent implements multi-functionality by enabling the robot cleaner to perform both cleaning operations and object detection/location functions using the same hardware platform. The cleaning map generation and object detection capabilities are integrated into the existing navigation system, allowing the robot to serve multiple purposes without requiring entirely separate systems.
Solution Approach 2:
The patent introduces a mobile terminal as an intermediary device that receives cleaning map data and object location information from the robot cleaner. This intermediary handles the complexity of data processing and presentation, allowing the robot to maintain relatively simple onboard processing while still providing sophisticated object detection and location capabilities through the combined system.
3Ease of operation
If robot cleaners do not provide object location information, then the ease of operation is high, but the user convenience is low
Solution Approach 1:
The patent implements feedback by transmitting object location information from the robot cleaner to the user's mobile terminal. The cleaning map with marked object positions is sent back to the user, providing immediate feedback about detected objects and their precise locations. This enables users to quickly find lost items without manually searching the entire cleaning area.
Data Source
AI summary
An artificial intelligence robot includes a camera configured to acquire image data, a memory configured to store an object recognition model used to recognize an object from the image data, and a processor configured to acquire a speech command, determine whether an intention of the acquired speech command is object search, recognize an object from the image data based on the object recognition model during traveling when the intention of object search, and output a notification indicating that the object has been recognized when the recognized object is an intended object.