Robotic Task Feasibility Analysis Using Semantic Maps and Ontologies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic task execution systems fail to assess the feasibility of tasks in uncertain, dynamic environments, leading to increased execution time and power consumption due to the lack of ability to check if user-provided instructions are executable.
Innovation Solution
A method and system for task feasibility analysis with explanation, utilizing a camera, odometer, and ontology to create semantic maps and global knowledge graphs, processing image frames to determine task feasibility and generate explanations for non-feasible tasks, enabling the robot to execute navigation and manipulation tasks efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the robot executes tasks without feasibility analysis in uncertain environments, then task execution can proceed without preliminary checks, but execution time and power consumption increase due to inability to avoid dead-end situations
Solution Approach 1:
The system performs task feasibility analysis before executing the actual task. The processor analyzes whether a task can be successfully completed by checking if required objects exist in the environment using semantic maps and knowledge graphs, preventing the robot from attempting impossible tasks and wasting time
2Productivity
If the robot attempts to execute all user-provided tasks without feasibility checking, then all tasks are processed, but power consumption increases due to executing non-feasible tasks
Solution Approach 1:
The system performs task feasibility analysis before executing the actual task. The processor analyzes whether a task can be successfully completed by checking if required objects exist in the environment using semantic maps and knowledge graphs, preventing the robot from attempting impossible tasks and wasting energy
Solution Approach 2:
The robot autonomously evaluates its own task feasibility using its built-in sensors, semantic maps, and knowledge graphs. The system self-determines whether tasks are executable based on environmental understanding without external assistance, optimizing energy usage by avoiding futile attempts
3Adaptability or versatility
If the robot lacks environmental knowledge for task planning, then task execution can proceed without complex processing, but the robot fails to adapt to unpredictable changes and exceptional scenarios
Solution Approach 1:
The system pre-builds semantic maps of the environment and knowledge graphs containing object relationships and attributes before task execution. This preliminary environmental understanding enables the robot to adapt to changes and exceptional scenarios by querying pre-established spatial and semantic knowledge
Solution Approach 2:
The system introduces semantic maps and knowledge graphs as intermediary data structures between the physical environment and task execution. These intermediaries encode environmental knowledge, object relationships, and spatial information, allowing the robot to reason about feasibility without direct complex environmental interaction
4Loss of information
If the robot executes tasks without providing explanations for failures, then system simplicity is maintained, but user understanding of task failure reasons is lost
Solution Approach 1:
The system provides feedback to the user by generating explanations for task failures. When a task is determined to be non-feasible, the system communicates the reason (e.g., required object not found in environment) back to the user, enabling better task formulation and system-human collaboration
Data Source
AI summary
This disclosure relates generally to a method and system for task feasibility analysis with explanation for robotic task execution. Conventional methods for task feasibility analysis does not utilize an ontology for task capability understanding. The present disclosure uses an explainable semantic approach for checking task feasibility in a real world. The method creates scene graphs which is further used for generating a global knowledge graph and a semantic map. These are used for task feasibility analysis for an input task instruction received from a user. When the user provides the task instruction the method checks whether it is feasible or not. This helps in avoiding dead end tasks and provides the user to alter the task instruction towards feasible task. The disclosed method is used for robotic task execution in an environment.