Context-Aware Reasoning and Interactive Decision-Making for Human-Robot Collaboration
Not stated
- Funding
- Self-Funded PhD Students Only
- Application deadline
- Year-round applications
About the project
About the Project As autonomous mobile robots become increasingly integrated into industrial, logistics and service environments, effective human-robot collaboration requires robots to make intelligent decisions that account for both environmental context and human requirements. While recent advances in artificial intelligence have significantly improved robot perception, navigation and natural language understanding, many robotic systems remain reactive, executing predefined tasks without considering changing contextual information or engaging users when uncertainty arises. Enabling robots to reason about their surroundings, interact naturally with people and adapt their behaviour in dynamic environments remains a significant challenge in intelligent autonomous systems. This project aims to develop a novel framework for context-aware reasoning and interactive decision-making that enables autonomous mobile robots to collaborate naturally and effectively with human users. Rather than treating decision-making as a fixed planning problem, the proposed research will investigate how robots can continuously integrate information from multimodal perception, environmental context and human interaction to support adaptive and explainable behaviour. The research will investigate methods for combining visual perception, natural language understanding, contextual reasoning and autonomous planning within a unified decision-making architecture. The robot will interpret high-level human requests, understand the surrounding environment and select appropriate actions while considering factors such as environmental changes, user preferences, safety constraints and task priorities. When information is incomplete or ambiguous, the robot will engage in interactive dialogue to request clarification, confirm assumptions or propose alternative courses of action, creating a collaborative decision-making process between human and robot. The project will investigate methods for modelling and representing contextual knowledge, incorporating temporal and spatial information, human intention, environmental constraints and operational objectives into the robot's decision-making process. Experimental validation will be conducted using indoor and outdoor mobile robotic platforms in representative collaborative scenarios. Human operators will interact with the robots using natural language to assign tasks, modify objectives and respond to changing operational conditions. The expected outcome is a novel context-aware decision-making framework that enables autonomous mobile robots to reason about complex situations, collaborate effectively with human users and adapt their behaviour in dynamic environments.