Moving Robot Structured Light Obstacle Detection for Traversal Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional moving robots rely solely on physical sensing values like height and distance to detect obstacles, which limits their ability to accurately recognize the shape and characteristics of obstacles, potentially leading to dangerous situations or restrictions.
Innovation Solution
The moving robot employs structured light irradiated in a predetermined pattern to detect obstacles while traveling, using image capturing and deep learning to analyze changes in the structured light over time, allowing for accurate obstacle recognition and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical sensing values (height and distance) are used to detect obstacles, then the detection process is simple, but the obstacle recognition accuracy is insufficient
Solution Approach 1:
The patent transitions from 1D physical sensing (height and distance measurements) to 2D/3D visual sensing by projecting structured light patterns and capturing images. The structured light projection system emits patterned light that reflects off obstacles and is captured by an image sensor, creating a two-dimensional visual representation that contains rich spatial information about obstacle shape, size, and position, thereby significantly improving recognition accuracy.
Solution Approach 2:
The patent introduces structured light patterns as an intermediary medium between the detection system and the obstacle. By projecting known light patterns onto the environment and analyzing their reflection and distortion, the system can infer obstacle characteristics without direct contact or complex sensors. The light patterns serve as a mediator that carries information about the obstacle's geometry and position back to the detector.
2Reliability
If only height-based obstacle determination is used, then the control logic is simple, but the robot may encounter dangerous situations or be restrained by obstacles
Solution Approach 1:
The patent performs preliminary obstacle classification by analyzing the shape, size, and visual characteristics of detected obstacles before the robot commits to a traversal action. The system uses image processing to identify obstacle types (e.g., thresholds, steps, furniture) and predicts potential dangers in advance, allowing the robot to select appropriate traversal strategies proactively rather than reactively, thereby improving safety and reducing restraint incidents.
Solution Approach 2:
The patent implements a feedback mechanism where the robot continuously monitors obstacle characteristics during traversal attempts. If the robot encounters resistance or detects that an obstacle is more complex than initially classified, the system adjusts its traversal strategy in real-time based on sensor feedback, preventing dangerous situations and reducing restraint by adapting to actual obstacle properties during the traversal process.
3Productivity
If the robot avoids all detected obstacles, then the robot safety is improved, but the cleaning productivity decreases due to unnecessary path deviations
Solution Approach 1:
The patent applies different traversal strategies to different types of obstacles based on their local characteristics. Instead of uniformly avoiding all obstacles, the system classifies obstacles by type (e.g., low thresholds, high obstacles, furniture) and selects appropriate traversal methods for each category. This localized approach allows the robot to traverse safe, traversable obstacles efficiently while avoiding genuinely dangerous obstacles, thereby improving both productivity and reliability simultaneously.
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
This approach enables the moving robot to quickly and accurately determine obstacles, avoid dangerous situations, and perform corresponding motions to navigate around or over obstacles, thereby improving cleaning efficiency and reducing the risk of restraint.
Implementation Method 1
an obstacle detector for capturing an acquired image with respect to a light pattern irradiated in a traveling direction and detecting an obstacle located in the traveling direction
Data Source
AI summary
In a moving robot and a control method according to the present disclosure, an obstacle is detected using structured light irradiated in a predetermined type of light pattern in a traveling direction while traveling, and a specified operation is performed in response to the obstacle. Moreover, a dangerous obstacle is recognized by extracting changes over time using a plurality of images for an obstacle or a low obstacle that is difficult to determine as detected data, and thus, it is possible to improve accuracy according to the determination of the obstacle, improve a corresponding operation according to the obstacle, minimize the uncleaned area while preventing restraint due to the obstacle, and improve the cleaning performance.