Distance Histogram Imaging for Cage-Free Robot Safety Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous agent systems require static safety cages and expensive on-AR sensors to ensure safe operation, which hinder close collaboration with humans and are limited by blind spots.
Innovation Solution
A cloud/edge-based system processes distance information from infrastructure sensors to generate histogram images, separating static and dynamic elements using distribution-based outlier analysis, enabling safe operation without static safety cages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static safety cages are erected around autonomous robots, then safety is improved, but collaboration with human co-workers deteriorates due to physical barriers
Solution Approach 1:
The patent replaces the mechanical safety cage system with an optical sensing system using infrastructure-mounted cameras and computer vision algorithms. The system processes video feeds to detect dynamic objects and generate occupancy grids, eliminating the need for physical barriers while maintaining safety through virtual monitoring zones.
Solution Approach 2:
The patent introduces an intermediary safety system consisting of infrastructure sensors and processing circuitry that mediates between autonomous robots and human workers. This intermediary layer detects potential collisions and triggers safety protocols without requiring direct physical separation, enabling closer human-robot collaboration.
2Difficulty of detecting and measuring
If on-AR sensor systems are installed on autonomous robots, then safety detection capability is improved, but system cost deteriorates due to expensive safety companions
Solution Approach 1:
The patent merges the safety sensing function into the existing infrastructure by mounting sensors on fixed structures rather than on each autonomous robot. Multiple robots share the same infrastructure sensor system, reducing per-robot costs while maintaining comprehensive safety coverage through centralized processing.
Solution Approach 2:
The infrastructure sensor system serves multiple functions simultaneously: it provides safety monitoring for multiple autonomous robots, tracks human workers, and generates occupancy grids for navigation. This multi-functional approach eliminates the need for dedicated safety companions on each robot.
3Speed
If on-AR sensor systems are used for safety monitoring, then real-time detection is improved, but operational reliability deteriorates due to blind spots
Solution Approach 1:
The patent positions sensors on infrastructure structures at strategic locations to achieve optimal coverage of specific zones. Each sensor is placed to monitor particular areas where autonomous robots operate, creating localized high-quality detection zones that collectively cover the entire workspace without blind spots.
Solution Approach 2:
The patent transitions from robot-centric sensor placement to infrastructure-centric sensor placement, changing the spatial dimension of safety monitoring. By mounting sensors on fixed structures rather than on moving robots, the system achieves stable, multi-angle coverage that eliminates blind spots while maintaining real-time detection capabilities.
Data Source
AI summary
An apparatus, including an interface configured to receive images with measured distance information of an environment in which an autonomous agent is designed to operate; and processing circuitry that is configured to: generate distance histogram images over time, wherein the distance histogram images include the measured distance information in corresponding picture elements; perform a distribution-based outlier analysis on the distance histogram images to classify each picture element of each of the received images as either an outlier picture element representing a dynamic object or a non-outlier picture element representing a static portion of the environment; and track a dynamic object over time and cause an action by the autonomous agent if it is determined, based on a result of the distribution-based outlier analysis, a distance between the dynamic object and the autonomous agent is less than a predefined distance.