Robot Visual Marker Detection for Dynamic Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collision avoidance methods for robots, such as using infrared sensors or laser rangefinders, are ineffective for dynamic obstacles like moving robots due to blind spots and limited sensitivity, and lack sufficient response time for collision prevention.
Innovation Solution
Robots are equipped with an identifying image and an image capturing device to detect and recognize another robot's identifying image, stopping movement when detected, and resuming when the other robot is stationary or taking a detour based on identification code comparison.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If infrared sensors or laser rangefinders are used to detect obstacles, then static obstacles can be detected effectively, but dynamic obstacles like moving robots cannot be detected reliably due to blind spots and limited sensitivity
Solution Approach 1:
The patent replaces traditional mechanical sensing systems (infrared sensors, laser rangefinders) with a visual detection system using image capturing devices. This substitution enables the robot to detect dynamic obstacles effectively by capturing images and identifying identifying images on other robots, overcoming the blind spots and limited sensitivity of mechanical sensors.
Solution Approach 2:
The patent uses image capturing devices to create visual copies (images) of the environment and other robots. By capturing and analyzing images containing identifying images, the system can detect and respond to dynamic obstacles without direct physical contact or reliance on limited-range sensors.
2Reliability
If traditional sensors are used for collision avoidance, then detection can be performed, but response time is insufficient for preventing collisions with dynamic obstacles
Solution Approach 1:
The patent implements preliminary detection by continuously capturing images and identifying identifying images of other robots before collision occurs. The system processes images to determine relative positions and predicts potential collision paths in advance, allowing the robot to take preventive action (change path, stop, or adjust speed) before the collision would occur, thus reducing response time loss.
3Productivity
If robots move along predetermined paths independently, then task performance is maintained, but collision risk with other robots increases
Solution Approach 1:
The patent implements a feedback mechanism where the robot continuously captures images, identifies identifying images of other robots, determines relative positions, and adjusts its movement based on this information. The processor controls the driving device to modify the predetermined path when potential collisions are detected, while maintaining task performance by resuming the original path after safe passage is confirmed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Effectively prevents collisions among robots by ensuring safe movement and maintaining task performance, such as lawn mowing or cleaning, by accurately identifying and responding to dynamic obstacles.
Implementation Method 1
an image capturing device, a storage device that stores a predetermined path, and a processor that is electrically connected to the driving device, the image capturing device and the storage device
Data Source
AI summary
A method for preventing a robot from colliding with another robot that is provided with an identifying image includes steps of: controlling the robot to move along a predetermined path; stopping the robot when it is determined that a first image captured by the robot while the robot is moving contains the identifying image; and controlling the robot to resume moving along the predetermined path when it is determined that a second image captured by the robot while the robot is not moving does not contain the identifying image.