Collaborative Robot Motion Planning With Vision-Based Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional collaborative robotic systems are either costly or limited in speed and payload, leading to inefficiencies and safety concerns when working alongside human laborers, as they either require expensive safety equipment or operate at reduced capabilities to avoid collisions.
Innovation Solution
A collaborative robotic assembly using mid-tier consumer-grade processors and non-point cloud cameras to estimate human joint positions in real-time, allowing for simultaneous and safe human-robot interaction by dynamically adjusting robotic motion plans based on live data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional collaborative robotic systems are used to allow simultaneous human-robot work, then productivity is improved, but safety hazards increase due to accidental bumping and collisions
Solution Approach 1:
The patent replaces expensive mechanical safety systems (caging, fencing, force-torque sensors, high-performance industrial computers) with a vision-based system using non-point cloud cameras and consumer-grade processors. The camera system captures visual data to detect human positions and predict collision risks, substituting mechanical safety infrastructure with optical sensing and software-based collision risk assessment
Solution Approach 2:
The patent introduces visual data from cameras as an intermediary between the robotic system and human workers. This visual information serves as a mediator that enables the robotic system to perceive human presence and adjust its behavior accordingly, facilitating safe collaboration without requiring direct physical interaction or expensive safety infrastructure
2Object-affected harmful factors
If expensive safety equipment such as LIDAR, stereo cameras, and high-performance industrial computers are deployed, then safety is improved, but system cost becomes prohibitive
Solution Approach 1:
The patent employs non-point cloud cameras and consumer-grade processors instead of expensive LIDAR systems, stereo camera arrays, and high-performance industrial computers. These cheaper components are sufficient for the application needs, reducing system cost while maintaining adequate safety monitoring capability through visual data processing
Solution Approach 2:
The patent extracts only the essential safety monitoring functionality needed for collaborative robotics, removing unnecessary expensive components. By using non-point cloud cameras and consumer-grade processors, the system retains collision risk detection capability while eliminating the need for costly high-performance hardware
3Object-affected harmful factors
If robotic systems are segregated from human laborers to ensure safety, then safety is improved, but productivity decreases due to sequential task execution
Solution Approach 1:
The patent makes the robotic system dynamic by continuously adjusting its motion plan based on real-time visual data of human positions. The system transitions from static segregated operation to dynamic collaborative operation, where the robotic implement adapts its behavior in response to detected human presence and predicted collision risk, enabling both safety and simultaneous task execution
Data Source
AI summary
A robotic system to facilitate simultaneous human laborer and robotic tasks on an article. The system includes data acquisition from a non-point cloud camera and implementation by a mid-tier consumer grade workstation. Nevertheless, a motion plan may be carried out by the robotic aid in a manner that allows for “on-the-fly” adjustment to a second motion plan to avoid collision with the laborer during the performed tasks.