Industrial Robot Collision Avoidance Using Common Figure Space
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current industrial robot systems lack effective collision detection and avoidance capabilities, particularly when collaborating with humans, as they rely on torque or force sensors, which only detect collisions after they occur, and camera or laser systems are prone to errors due to lighting conditions and require extensive calibration and multiple sensors, leading to reduced dynamics and increased costs.
Innovation Solution
A method and system that utilize an environment detection unit and position detection unit to transform data into a common figure space, creating a control figure and object figure, which account for temporal and spatial correlations, movement history, and relative velocities to generate action instructions for the industrial robot, enabling early detection and avoidance of potential collisions through a parameter set and predefined criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera systems are used for person detection, then detection capability is improved, but sensitivity to lighting conditions causes errors in image recognition
Solution Approach 1:
The system segments the detection task by using multiple camera systems operating at different wavelengths (visible light and infrared), where each camera type detects different aspects of the environment, thereby compensating for the limitations of individual camera systems under varying lighting conditions
Solution Approach 2:
The system combines data from different camera types (visible light cameras and infrared cameras) to create a composite detection result, similar to using composite materials, where each component contributes its strengths to overcome the weaknesses of individual components
2Area of stationary object
If multiple external cameras are installed to ensure reliable observation, then detection coverage is improved, but device complexity and cost increase
Solution Approach 1:
The robot system integrates detection functions directly into its existing structure, allowing the robot's body and components to serve dual purposes: both performing manufacturing tasks and acting as detection platforms, thereby eliminating the need for separate external camera installations
Solution Approach 2:
The system merges the detection functionality with the robot's existing structure and components, combining multiple functions into a unified system rather than adding separate external detection devices
3Adaptability or versatility
If camera systems are mounted on the industrial robot, then detection follows the robot movement, but shading patterns reduce detection reliability
Solution Approach 1:
The system segments the detection task across multiple wavelengths (visible and infrared), where infrared detection compensates for shading issues in visible light, allowing the robot to maintain detection capability while moving through shaded areas
Solution Approach 2:
The system uses infrared radiation as an intermediary detection method that can penetrate or bypass shading patterns that block visible light, providing a complementary detection channel that maintains reliability during robot movement
4Device complexity
If torque or force sensors are used for collision detection, then simplicity is maintained, but collision detection only occurs after actual collision
Solution Approach 1:
The system performs preliminary detection of people and objects in the robot's path using vision systems before the robot executes movements that could lead to collision, allowing the control system to plan alternative paths or slow down in advance, thereby preventing collisions rather than detecting them after occurrence
Solution Approach 2:
The system continuously feeds back detection information from multiple camera systems to the control unit, which adjusts robot movements in real-time based on detected people and objects, creating a closed-loop system that prevents collisions through continuous monitoring and adaptive control
Data Source
AI summary
A method for controlling an industrial robot provides that position data of the industrial robot are detected and environment data of an object in an environment of the industrial robot are captured with an environment detection unit. The position data and the environment data are transformed into a common figure space, in which a control figure is defined for the industrial robot and an object figure of the object is represented. A parameter set is created which takes a dimensioning of the control figure in the figure space into account. The parameter set comprises a temporal and spatial correlation of the position data and the environment data and takes into account the movement history of the industrial robot and/or of the object. An action instruction is generated for the industrial robot if the control figure and the object figure satisfy a predefined criterion in relation to each other.