Abstract:Knowledge about how well a robot can perform a specific task is currently present only in engineering reports which are inaccessible to the robot. Artificial Intelligence techniques, such as hypergraphs and automated reasoning, can provide such engineering knowledge online while enabling updates in the knowledge with new experiences. This requires a sound knowledge structure and maintenance routines for keeping this knowledge-base about the robot's capabilities truthful. A robot with such up-to-date information can reason about if and how well it can accomplish a task. This article introduces a knowledge representation that combines an ontology on system engineering, a deductive reasoning on the connections between system components, and an inductive reasoning on the performance of these components in the current system configuration. This representation is further used to derive the expected performance for the overall system based on a continuous evaluation of the actual performance per component. Our real-life implementation shows a robot that can answer questions on whether it can do a specific task with the desired performance.

What problem does this paper attempt to address?

### What problem does this paper attempt to solve? This paper aims to solve the problem of how a robot system can evaluate its own capabilities and select the most appropriate configuration to complete tasks when performing tasks in complex and unpredictable "open - world" environments. Specifically, the paper proposes a method named Self - X. By constructing and maintaining an updatable knowledge base, robots can evaluate the performance of their configurations online and select the optimal configuration according to the current operating conditions. #### Main problems: 1. **Self - evaluation ability**: Currently, knowledge about robot performance only exists in engineering reports, which are inaccessible to the robots themselves. Therefore, robots cannot judge by themselves whether they can successfully complete specific tasks. 2. **Adaptability to dynamic environments**: As robots expand from closed environments (such as work cells) to more complex open environments (such as factories, homes), they need to have stronger self - adaptive abilities to cope with changing task requirements and environmental conditions. 3. **Multi - task processing**: Future robots not only need to perform a single task, but also need to be able to handle a series of different types of tasks, even new tasks that were not foreseen in the early design stage. #### Solutions: - **Knowledge representation**: A knowledge representation method that combines system engineering ontology, deductive reasoning of connections between components, and inductive reasoning of component performance is introduced. This representation can continuously evaluate the actual performance of each component, thereby inferring the expected performance of the entire system. - **Adaptive configuration selection**: Based on the above knowledge representation, a mechanism is developed to enable robots to select the most suitable configuration to complete tasks according to the current operating conditions. This includes identifying which available configurations match the current conditions and which configurations can meet the performance requirements of the tasks. - **Real - time update**: The operating conditions and system performance are evaluated in real - time through an online module to ensure that the information in the knowledge base is always up - to - date, so that robots can make accurate self - evaluations and decisions. #### Key contributions: 1. **Ontology design**: An ontology for system engineering knowledge that can be used at runtime is created. This ontology not only describes the design of hardware and software components, but also covers their functions and relationships. 2. **Hypergraph implementation**: The ontology is implemented as a hypergraph structure, allowing the performance data of internal components to be updated with actual information. 3. **Reasoning methods**: Deductive and inductive reasoning methods are developed to analyze the effectiveness of system configurations. Through these methods, Self - X enables robots to more intelligently evaluate their execution capabilities and select the best action strategies when facing new tasks and unknown environments.

Can I do it

A User Interface for Sense-making of the Reasoning Process while Interacting with Robots

A survey of ontology-enabled processes for dependable robot autonomy

RoboBrain: Large-Scale Knowledge Engine for Robots

Robot Planning Based on Behavior Tree and Knowledge Graph

Integrated Commonsense Reasoning and Deep Learning for Transparent Decision Making in Robotics

What you need to know about a learning robot: Identifying the enabling architecture of complex systems

Hierarchical Understanding in Robotic Manipulation: A Knowledge-Based Framework

Knowledge-Driven Robot Program Synthesis from Human VR Demonstrations

Deliberative and Conceptual Inference in Service Robots

Robot Representation and Reasoning with Knowledge from Reinforcement Learning

A Semantic Approach for Building System Operations: Knowledge Representation and Reasoning

Intelligent Medical Knowledge Robot Based on Knowledge Graph

A Cognition Knowledge Representation Model Based on Multidimensional Heterogeneous Data

An Experimental Autonomous Articulated Robot that can Learn

Ontology based autonomous robot task processing framework

A Survey of Knowledge Representation in Service Robotics

Multiple Test Analyzer (MTA)

Graph-based Hierarchical Knowledge Representation for Robot Task Transfer from Virtual to Physical World

A Knowledge Driven Approach to Adaptive Assistance Using Preference Reasoning and Explanation

A graph-based spatial temporal logic for knowledge representation and automated reasoning in cognitive robots