Service Robot Task Control for New Situation Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Service robots lack flexibility in handling new situations and require constant human intervention for operation, limiting their efficiency and usability.
Innovation Solution
A service robot system with a processing unit that determines tasks, uses sensors to identify the environment, evaluates action definition candidates based on past success scores, and allows for remote operator assistance when needed, enabling autonomous task execution and learning from user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the service robot operates autonomously based on pre-defined rules, then the robot can perform basic tasks independently, but the robot lacks flexibility when encountering new situations and requires human intervention
Solution Approach 1:
The patent implements a feedback mechanism where the robot captures images of its environment, transmits them to a server, receives evaluation results comparing current situations with historical data, and uses this feedback to improve its autonomous decision-making. The server analyzes situation similarity scores and provides guidance back to the robot, enabling it to adapt to new situations while maintaining autonomous operation.
Solution Approach 2:
The system pre-stores multiple action definition candidates in the database before the robot encounters new situations. When the robot faces a situation, it retrieves pre-stored action definitions and evaluates them against current conditions. This preliminary preparation of action definitions enables the robot to respond flexibly to new situations without requiring real-time human intervention.
2Adaptability or versatility
If the operator remotely controls the robot in real-time, then the robot can handle new situations with human intelligence, but the system requires constant human intervention and reduces operational efficiency
Solution Approach 1:
The patent implements partial remote intervention where the operator is only involved when the situation similarity score falls below a threshold. For routine situations with high similarity scores, the robot operates autonomously without operator input. This partial intervention approach maintains adaptability for new situations while preserving operational efficiency for common tasks.
Solution Approach 2:
The robot performs self-evaluation of its situations by comparing current environmental data with historical situation databases. It autonomously determines whether a situation is familiar or new based on similarity calculations, and only requests human assistance when necessary. This self-service capability reduces constant human intervention while maintaining the ability to handle new situations.
3Reliability
If the robot stops operation to remain in a safe state when encountering unresolved situations, then the robot ensures safety, but the operational efficiency and continuity are reduced
Solution Approach 1:
The system pre-stores multiple action definition candidates and safety protocols in the database before the robot encounters unresolved situations. When a new situation is detected, the robot immediately retrieves pre-prepared action definitions and safety procedures, allowing it to respond without halting operation. This preliminary preparation enables continuous operation while maintaining safety through pre-planned responses.
Solution Approach 2:
The robot continuously monitors situation similarity scores and receives real-time feedback from the server about appropriate actions for new situations. This continuous feedback loop allows the robot to adapt its behavior dynamically without stopping, maintaining both safety and operational continuity by adjusting actions based on incoming information rather than halting entirely.
Data Source
AI summary
The invention regards a service robot system comprising a robot and a learning method for the system. The robot has a drive system and at least one effector. The system's processing unit determines a task to be executed by the robot and controls the drive system and the effector according to the task. The processing unit automatically retrieves action definition candidates from a database, evaluates the retrieved action definition candidates with respect to a success score indicating a likelihood that an action according to the action definition candidate contributes to successfully fulfilling the task, executes an action according to the action definition candidate having the highest probability equal to or above a predefined threshold, and sends a request for assistance via a communication interface if the success score is less than a preset threshold. The system will learn from instructions and additional information received in response to such request.