AreaGoal Robot Navigation with Semantic Branch Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing telepresence systems face challenges in ease of use for non-expert users, particularly in unfamiliar environments, requiring continuous manual intervention for navigation and are hardware-dependent, limiting software enhancements.
Innovation Solution
A modular navigation system using 'PointNav' and 'AreaNav' modules for robot navigation, incorporating RGB-D sensors, odometry, and semantic maps, enabling speech-based interaction and decoupled path planning to navigate from one area to another, with a focus on 'AreaGoal' tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual navigation capability is provided in existing telepresence systems, then the robot can move within the environment based on remote operator's command, but the operation becomes cumbersome for non-expert users in non-familiar environments requiring continuous user intervention
Solution Approach 1:
The system enables autonomous navigation where the robot independently plans and executes navigation paths from one area to another using semantic maps and path planning algorithms, eliminating the need for continuous manual control by the remote operator while maintaining the ability to perform navigation tasks autonomously
2Adaptability or versatility
If existing telepresence systems are tightly coupled with robot hardware, then the system can utilize hardware resources, but software enhancement becomes difficult particularly by third-party developers
Solution Approach 1:
The system is divided into independent modular components including semantic map generation module, path planning module, and navigation execution module, where each module can be independently developed, modified, or replaced by third-party developers without affecting the entire system, thereby enabling flexible software enhancement while reducing hardware coupling
Solution Approach 2:
The navigation system is designed with universal interfaces and standardized data structures that allow different software implementations to work with the same hardware platform, enabling third-party developers to create enhanced navigation algorithms without being constrained by hardware-specific implementations
3Productivity
If manual navigation requires continuous user intervention to move the remote robot from place to place, then the robot can navigate within the remote environment, but the user intervention becomes time-consuming and reduces navigation efficiency
Solution Approach 1:
The system pre-generates semantic maps of the environment and pre-plans navigation paths before execution, allowing the robot to quickly navigate from one area to another without requiring real-time continuous user intervention, thereby reducing navigation time and increasing efficiency
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for navigation of a robot from a first area to a second area in a facility is provided. The present disclosure is providing robot navigation using the 'Areagoal' Navigation technique. 'Areagoal' class of problem is divided into two subtasks: identifying the area; and navigation from one area to another. The robot starts in first location and goes out of the current area if it is not in the target area. If there are multiple openings from the first area, it needs to select the most statistically close one to the target area and go there. If the target area is not reached, it backtracks to an earlier viable branch position to continue the target area search. The system takes input from RGB-D camera and odometer, while the output is action space (left, right, forward) with goal of moving to target area.