TALOS was featured in a recent episode of the EU Energy Projects Podcast, hosted by Enlit World, with project coordinator Daniel Albuquerque (EDP) previewing the consortium’s progress as the initiative enters its final months.
Albuquerque pointed to the sector backdrop driving the project’s relevance: PV operations are growing more complex as plants move into more remote locations, O&M costs rise, and safety constraints and workforce limitations tighten. TALOS, he said, tackles a concrete, well-defined problem within that shift.
One highlight of the three-year project, now in its closing stages, has been an open call that funded European SMEs and startups to broaden its scope, a co-creation approach Albuquerque described as unusual in his experience. The result: 30 additional solutions delivered on top of those initially proposed, across TALOS’s three target PV configurations of ground-mounted, floating, and agrivoltaic installations.
Panel cleaning illustrates the range of hardware involved. For floating PV, the consortium built an approximately 80kg robot roughly the size of a domestic robotic vacuum cleaner. For utility-scale ground-mounted installations, a 12-tonne machine fitted with large brushes and a water tank does the job instead.
AI runs underneath the hardware, drawing on multiple datasets, including site mapping, to support cleaning, panel defect detection, and, in agrivoltaic settings, crop health and yield monitoring. Most importantly, Albuquerque noted, those datasets are what let the robots operate autonomously: a challenge he likened to the decade-plus Tesla has spent pursuing full self-driving, complicated further in a power plant setting by operational risks and, in remote locations, unreliable internet connectivity.
In its remaining months, TALOS is demonstrating its full ecosystem of solutions across all three PV scenarios, alongside technological, economic and environmental analyses meant to mature the technologies toward market-ready solutions for PV operators.
Source: “How TALOS is revolutionising PV management with robotics,” Enlit World, 27 May 2026
