NOVEA is an independent engineering consultancy for deep-tech teams solving hard, cross-disciplinary problems in space, defense, medtech, and robotics.
NOVEA owns the full stack, from concept to production. Where a problem would normally need a team of specialists across control, software, hardware, and vision, NOVEA closes it as a single coherent engineering effort. The result: less coordination, faster progress, lower technical risk, and less costly trial and error.
Vincent Mistler
Founder · GNC & Dynamical Systems Engineer
Vincent Mistler has spent 25+ years turning hard engineering into working systems across aerospace, defense, automotive, medtech, and robotics, much of it in mission-critical and certified environments.
His work began in flight control and autonomy. In 2001, he published an early and widely cited control formulation for quadrotor UAVs with the IEEE, foundational work from a time when multirotor flight was barely explored. At Safran, he developed flight-control and avionics systems for helicopters under DO-178 certification, including autopilots for search-and-rescue aircraft and vision sensors for military rotorcraft. At Delphi, he built advanced control systems for automotive emissions reduction.
At SES, one of the world's largest satellite operators, he provided GNC and flight dynamics expertise across the fleet: attitude control and propulsion, maneuver planning and execution, acquisition and commissioning of new satellites, fleet performance analysis, and resolution of in-orbit anomalies.
Through NOVEA, he works with deep-tech teams on the problems that sit between disciplines: GNC and relative navigation for spacecraft, vision-based estimation and perception, sensor fusion, and the embedded software that carries all of it from the whiteboard to real hardware. Recent work spans on-board space situational awareness, autonomous rendezvous and proximity operations, vision-based drone detection and counter-UAV simulation, and end-to-end systems for advanced microscopy.
Across every domain, the approach is the same: specify first, attack the riskiest unknowns early, build in small validated steps, and prove in real conditions.
