Robot Motion Planning Using Historical Human Activity Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional robot navigation systems in occupied environments are computationally intensive due to real-time sensing requirements, making them poorly suited for task and motion planning, and they struggle to efficiently navigate complex indoor spaces shared with humans.
Innovation Solution
A robot navigation system that utilizes a sensor network to generate historical human activity data, deriving path data to plan waypoints for tasks and adjusting movement based on real-time sensed data, combining global route planning with local reactive planning to optimize navigation in shared indoor environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real-time sensing is employed to detect human presence and enable robot navigation, then the robot can avoid contacting humans, but the computational resources of the device are heavily consumed
Solution Approach 1:
The system performs preliminary action by collecting historical human activity data in advance and generating pre-computed path data that represents optimal routes through the environment. This pre-processing of navigation information allows the robot to operate with reduced real-time computational requirements, as the heavy path planning computations are performed beforehand rather than during actual navigation
Solution Approach 2:
The navigation system is segmented into multiple components: a server that performs complex path planning using historical data, and a robot that executes pre-computed paths with minimal real-time processing. This segmentation separates the computationally intensive tasks from the real-time navigation tasks, allowing each component to operate efficiently within its designated function
2Ease of operation
If real-time sensing and react approaches are used for immediate human avoidance, then the robot can respond to immediate surroundings, but the approach is poorly suited for overall task and motion planning
Solution Approach 1:
The system segments navigation into two distinct layers: global path planning that uses historical data to determine optimal routes for overall task completion, and local reactive planning that handles immediate obstacle avoidance. This segmentation allows each layer to specialize in its respective function without the complexity of integrating real-time sensing into every decision level
Solution Approach 2:
Pre-computed path data acts as an intermediary between historical human activity patterns and real-time robot navigation. This intermediary layer translates complex historical data into simplified navigation instructions that guide the robot through the environment while maintaining awareness of human activity patterns without requiring complex real-time analysis
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present approach relates to navigation (e.g., route planning and movement) of robots in an indoor environment shared with humans. The present approach includes detecting human activity over time, including but not limited to human motion; modeling human activities using the historical human activity, and using the modeled human activity to plan robotic motion or movement.