Evaluation of shared intent–objective–goal model improve coordination in heterogeneous human–ai–robot–sensor teams

Human–AI–robot–sensor teams are increasingly used in missions where conditions change quickly, information is incomplete & communications may be constrained.

Examples: emergency response, infrastructure inspection, environmental monitoring & security operations.

Such teams combine strengths: human judgement, AI-supported analysis, robotic mobility & persistent sensing.

However, they are difficult to coordinate because each participant may hold a different view of mission priorities, available information & permissible action. Current coordination relies on task allocation, predefined workflows, comms procedures & shared operational pictures. These approaches may be less effective when plans are disrupted. This project will investigate if a shared, explicit Intent–Objective–Goal (IOG) model can improve coordination in heterogeneous human–AI–robot–sensor teams. The model will provide a common representation of: -Intent: the mission purpose & priorities. -Objectives: the effects the team is trying to achieve. -Goals: the meeting of mission targets that contribute to those objectives. The proposed innovation is not to give team members the same task list. It is to provide humans & autonomous systems with a common, understandable structure showing why actions matter, selecting priorities, explaining decisions & adapting activity when the original plan no longer applies. The central question is: Under what conditions does a shared, explicit IOG model improve the effectiveness of heterogeneous human–AI–robot–sensor teams, compared with conventional coordination arrangements? Technical plan outline -The project will develop & evaluate an initial IOG-based coordination approach in representative mission scenarios.

1. Define an IOG model that can be understood by human operators while also being represented in a form usable by AI, robots & sensor systems. The model will link mission purpose, operational objectives, actionable goals, constraints, authority & evidence

2. We develop a prototype coordination layer or interface through which participating agents can access the current IOG, report relevant evidence & align or re-plan activity. Humans will retain oversite allowing them to set, revise & override mission intent & objectives. Autonomous sensors will contribute data, confidence & relevance to the IOG rather than being treated as holders of intent.

3. The IOG approach is audited against conventional approaches using representative mission scenario data. The comparison will retain the same team capabilities & information, only the coordination approach differs. Scenarios will include routine & disrupted operations. Evaluation will focus on if IOG-based coordination improves outcomes.

Innovation & novelty: The project does not aim to replace task-based autonomy, which is already in use. It investigates if the IOG approach improves cross-team coordination when task allocations or plans change.

There is limited data on this explicit IOG approach. This project aims to provide this data. It will identify where this model offers tangible value & where conventional approaches remain sufficient and what are the requirements for: -Safe & accountable deployment. -Mission success & quality. -Time & resource efficiency. -Safety & risk management. -Recovery & replanning following disruption. -Coordination & communication burden. -Human understanding, workload & appropriately calibrated trust. -Anticipated outcomes & impact.