This paper investigates how AI can support cooperative engineering workflows, using the European Rover Challenge (ERC) as a case study. ERC requires student teams to define requirements, design subsystems, and integrate a complex rover within a single academic year, under tight schedules and strong subsystem interdependence. We distributed a role‑adaptive 40‑question survey to ERC 2025 participants, obtaining 104 valid responses from 14 teams, covering team structure, knowledge transfer, task management, integration practices, communication patterns, and current AI usage.
The survey revealed recurring bottlenecks: sparse documentation, unclear requirements, fragmented communication, informal task monitoring, and extensive integration rework. From these observations we derived key AI‑augmented workflow requirements: (1) assistance for task and requirement clarification; (2) compliance and change tracking; (3) automatic summarization of communications; (4) integration risk detection; (5) continuous knowledge capture and reuse. In response, we propose an initial system architecture comprising a user‑facing interface layer, credential management, service orchestration, specialized AI services (e.g., natural‑language understanding, code generation, risk reasoning), and connectors to engineering tools such as CAD, simulation, and version control. The architecture aims to reduce information asymmetry, improve task visibility, and flag integration conflicts early, thereby accelerating the overall development cycle.
Review