Image-Based Position Specification for Autonomous Device Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous robot systems struggle to navigate to unregistered positions, limiting user control to only predefined destinations.
Innovation Solution
A method involving image capture and processing to identify corresponding positions using feature values, allowing users to specify arbitrary positions by capturing images with a terminal device, which then generates and transmits movement commands to the autonomous device based on map information and SLAM technology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If predefined destinations are used for autonomous robot navigation, then system simplicity is maintained, but navigation flexibility and adaptability deteriorate
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing feature values for multiple candidate positions in the environment map before the user specifies a destination. When the user captures an image at the desired location, the system can quickly compare the captured feature values against the pre-stored data to identify the target position, eliminating the need for real-time complex calculations and enabling flexible navigation to any location.
Solution Approach 2:
The system creates a copy of the environment represented as a map with stored feature values at various positions. Instead of navigating to arbitrary coordinates directly, the system captures images at candidate positions, extracts feature values, and stores these as representations of each position. This allows the autonomous robot to navigate to the copied representation (matching feature values) rather than dealing with complex continuous coordinate specifications.
2Adaptability or versatility
If arbitrary position specification is enabled, then navigation flexibility improves, but position identification accuracy deteriorates
Solution Approach 1:
The system replaces direct coordinate-based position specification with image-based feature value matching. Instead of requiring precise manual coordinate input or complex mechanical positioning systems, the user simply captures an image at the desired location, and the system automatically identifies the position by comparing feature values, achieving both flexibility and accuracy.
Solution Approach 2:
The system introduces feature values as an intermediary between the user's position specification intent and the autonomous robot's navigation execution. The feature values extracted from captured images serve as a mediator that bridges the gap between arbitrary user-specified locations and the robot's navigation system, enabling accurate position identification without requiring direct coordinate manipulation.
3Measurement precision
If feature value comparison is performed across multiple candidate positions, then position identification accuracy improves, but processing time increases
Solution Approach 1:
The system performs the time-consuming task of feature value extraction and storage in advance, during environment mapping. By pre-processing and storing feature values for all candidate positions before the user needs to specify a destination, the system eliminates the need for real-time feature extraction and comparison, significantly reducing the time required for position specification while maintaining high accuracy through comprehensive candidate evaluation.
Data Source
AI summary
A method for controlling an autonomous device that moves in two dimensions using a controller includes obtaining a first image at a first position, which is a destination of the autonomous device, calculating, from the first image, first feature values indicating certain characteristics of the first image, referring to map information indicating correspondences between coordinate information indicating coordinates of defined positions included in a movement area of the autonomous device and second feature values, which are calculated from second images and indicate certain characteristics of the second images and identifying, by referring to the map information, a second position corresponding to second feature values having at least a predetermined degree of correspondence to the feature values generating a command for moving the autonomous device to the second position on the basis of coordinate information corresponding to the second position, and transmitting the command to the autonomous device.