AI Mobile Robot Image-Based Obstacle Learning for Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobile robots face limitations in recognizing and responding to various obstacles during cleaning, often resulting in collisions and errors due to delayed obstacle recognition and limited ability to adapt to new environments, leading to inefficient navigation and potential damage.
Innovation Solution
A mobile robot equipped with an image acquirer and obstacle detection unit that captures images while traveling, analyzes obstacle features, and determines response motions based on pre-captured image data, allowing for immediate avoidance and adaptation to different obstacles through communication with a server for updated obstacle information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the mobile robot uses conventional obstacle detection devices (infrared rays or laser beams) to detect obstacles, then the robot can detect obstacles within a predetermined distance, but the robot responds too late and may collide with the obstacle causing damage
Solution Approach 1:
The mobile robot captures images continuously during travel before actually encountering the obstacle. The controller stores these image data and processes them to identify obstacles in advance, allowing the robot to prepare avoidance maneuvers before collision risk arises. This preliminary image capture and processing enables earlier obstacle recognition compared to conventional detection methods that only detect when the robot is already close to the obstacle.
2Reliability
If the mobile robot immediately changes path when an obstacle is detected, then the robot avoids collision, but the robot cannot clean the corresponding region and must change path again after approaching maximum distance
Solution Approach 1:
The controller identifies obstacles and determines appropriate avoidance strategies in advance by processing stored image data before the robot reaches the obstacle. This allows the robot to plan its path deviation optimally, avoiding unnecessary detours and ensuring it can return to clean the area after passing the obstacle, thereby maintaining both collision avoidance and cleaning efficiency.
3Productivity
If the mobile robot approaches the obstacle for cleaning, then the robot can clean the region, but the robot may collide with the obstacle causing damage to target objects
Solution Approach 1:
The robot captures and processes image data in advance to identify obstacles before approaching them for cleaning. The controller uses this pre-processed information to determine safe approach distances and timing, allowing the robot to clean near obstacles without colliding with them or the objects they may contain. This preliminary recognition enables precise control of the approach maneuver.
4Measurement precision
If the mobile robot uses a separate recognition module to recognize human feet, then the robot can easily recognize this specific obstacle, but the robot still has limited ability to recognize various types of obstacles
Solution Approach 1:
The mobile robot uses a universal image acquisition device and controller that can identify multiple types of obstacles (human feet, furniture, office supplies, etc.) through a single multi-functional system. The controller processes image data to recognize various obstacle types based on their visual characteristics, eliminating the need for separate specialized modules for each obstacle type while maintaining high recognition accuracy across diverse obstacles.
5Adaptability or versatility
If the mobile robot performs tests through various experiments before product release, then the robot can learn information on obstacles, but the robot cannot accumulate information on all obstacles due to restrictive test environment
Solution Approach 1:
The mobile robot autonomously captures images and processes them to identify obstacles during actual operation in diverse real-world environments. The controller stores and analyzes image data from various locations and conditions, enabling the robot to continuously learn and accumulate obstacle information autonomously without being constrained by controlled test environments. This self-learning capability allows the robot to adapt to new obstacle types encountered in practice.
Data Source
AI summary
An artificial intelligence (AI) mobile robot and a method of controlling the same for learning an obstacle are configured to capture an image while traveling through an image acquirer, to store a plurality of captured image data, to determine an obstacle from image data, to set a response motion corresponding to the obstacle, and to operate the set response motion depending on the obstacle, and thus, the obstacle is recognized through the captured image data, the obstacle is easily determined by repeatedly learning an image, and the obstacle is determined before the obstacle is detected or from a time point of detecting the obstacle to perform an operation of a response motion, and even if the same detection signal is input when a plurality of different obstacles is detected, the obstacle is determined through the image and different operations are performed depending on the obstacle to respond to various obstacles, and accordingly, the obstacle is effectively avoided and an operation is performed depending on a type of the obstacle.