3D Safe Motion Planning for Human-Robot Workspaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing industrial machinery safety systems struggle to efficiently integrate human movement into motion planning while adhering to stringent safety standards, often leading to suboptimal operation or unnecessary shutdowns due to the unpredictability of human movements and the complexity of 3D workspaces.
Innovation Solution
A system utilizing multiple 3D sensors (e.g., time-of-flight cameras, 3D LIDAR, stereo vision) to create a dynamic 3D representation of the workspace, identifying safe zones, and generating constrained motion plans that avoid unsafe areas, adjusting in real-time to ensure safe and efficient operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional guarding approaches (cages, light curtains) are used to ensure human safety, then safety is improved, but workspace utilization and machinery efficiency deteriorate due to unnecessary shutdowns and constrained operations
Solution Approach 1:
The system dynamically adjusts the robot's operational state (stop, slow down, continue) based on real-time detection of human presence and predicted trajectory. The safety constraints are not static but adapt to the evolving situation, allowing the robot to maintain efficient operation when safe while ensuring protection when humans approach
Solution Approach 2:
The system continuously monitors the workspace using 3D sensors to detect human presence, determines predicted human trajectories, and feeds this information back to the motion planning system. This closed-loop feedback enables real-time adjustment of safety constraints, resolving the contradiction between safety and efficiency
2Measurement precision
If 2D LIDAR sensors are used for guarding, then detection capability is improved, but safety coverage deteriorates because they cannot detect intrusions beyond a single plane (e.g., legs detected but arms closer to robot remain undetectable)
Solution Approach 1:
The system transitions from 2D LIDAR sensing to 3D sensing using either 3D LIDAR or time-of-flight cameras. This dimensional upgrade enables comprehensive monitoring of the entire workspace volume, detecting humans at all distances and angles, thereby achieving both precise measurement and complete safety coverage
3Reliability
If 3D sensors are used to detect human presence, then safety coverage is improved, but system complexity and computational requirements worsen due to the need for real-time 3D workspace modeling and trajectory prediction
Solution Approach 1:
The system performs preliminary actions by pre-computing multiple possible human trajectories and predicting future human positions before actual collisions occur. This proactive approach simplifies real-time decision-making by having safety constraints already prepared based on predicted scenarios
Solution Approach 2:
The system uses dynamic modeling to predict human movement patterns, transforming the complex 3D sensing problem into a more manageable prediction problem. By modeling human dynamics and typical movement behaviors, the system reduces computational complexity while maintaining high safety coverage
4Reliability
If motion planning incorporates real-time human detection and safety constraints, then safety is improved, but operation efficiency deteriorates due to frequent trajectory recomputations and shutdowns
Solution Approach 1:
The system pre-computes multiple candidate trajectories and predicts future human positions in advance. When a human is detected, the system can quickly select from pre-prepared safe trajectories rather than computing from scratch, significantly reducing the time loss while maintaining safety
Solution Approach 2:
The system applies safety constraints selectively based on the situation. When humans are detected in dangerous zones, full safety constraints are applied; when the workspace is clear, constraints are relaxed or removed entirely. This partial application of safety measures maintains efficiency while ensuring safety when needed
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
Enables safe and efficient operation of machinery by dynamically adapting to changing workspace conditions, maximizing machinery use while ensuring human safety by avoiding collisions and adhering to safety standards.
Implementation Method 1
3D time-of-flight cameras, 3D LIDAR
Implementation Method 2
light curtains that determine if any object has intruded into a region monitored by one or more light emitters and detectors
Implementation Method 3
stereo vision cameras
Data Source
AI summary
A method of safely operating machinery in a workspace includes recording images of a portion of a workspace. The method also includes generating a three-dimensional (3D) representation of the portion of the workspace based on the recorded images, where the 3D representation includes one or more volumes that correspond to the portion of the workspace. Additionally, the method includes identifying one or more of the volumes as being either occupied or unoccupied. Further, the method includes mapping one or more safe zones based on the one or more identified volumes, where the safe zones correspond to one or more regions within the portion of the workspace for safe operation of machinery.