Solar Impulse 2 Crashes Into Gulf After Skydweller Conversion

The aircraft that once made aviation history by circumnavigating the globe on sunlight alone is now at the bottom of the Gulf of Mexico. Solar Impulse 2, the record-setting solar plane built by Bertrand Piccard and André Borschberg, crashed during an autonomous test flight on May 4 after losing power. No one was on board, and there were no injuries.

The crash, reported by the National Transportation Safety Board, closes the operational chapter of one of the most ambitious experimental aircraft ever built. By the time it went down, it wasn’t really Solar Impulse 2 anymore. It was a Skydweller military surveillance prototype wearing the same airframe.

From Record-Setting Solar Plane to Military Surveillance Asset

Solar Impulse 2 made its name in 2016 as the first fixed-wing, fully solar-powered aircraft to circle the planet. The trip took 16.5 months and 17 stops, with Piccard and Borschberg alternating pilot duties at cruise speeds between 31 and 62 mph. The aircraft cruised on sunlight by day and stored battery power for night legs.

In 2019, the Solar Impulse Foundation sold the aircraft to Skydweller Aero, a Spanish-American company with a very different roadmap. Skydweller wasn’t chasing climate awareness. The company wanted to turn the solar plane into a long-endurance, uncrewed surveillance platform for military and commercial customers.

The pitch: a solar drone that could loiter for weeks or months over latitudes between Miami and Rio de Janeiro, carrying radar, optical sensors, signals intelligence gear, and communications relays at a fraction of what satellites cost. Leonardo, the Italian defense giant, came on as a technology partner and prime contractor for European and NATO markets.

What the Aircraft Became

The original Solar Impulse 2 had a 232-foot wingspan, weighed about 5,100 pounds with a carbon fiber frame, and carried 17,248 solar cells producing 66 kW of peak power. After Skydweller’s modifications, the converted aircraft grew to a reported 236-foot wingspan, 5,620 pounds, and roughly 100 kW of solar power from its updated cell layout. It also picked up nearly 1,400 pounds of lithium-ion batteries for night operations.

Skydweller completed its first uncrewed flight in Spain in 2023, then moved testing to Stennis International Airport in Kiln, Mississippi. By 2024 and 2025, the company was running progressively longer autonomous missions, including a 73-hour endurance flight for the U.S. Navy as part of the Autonomous Maritime Patrol Aircraft program.

Those tests positioned Skydweller for U.S. Southern Command missions: drug interdiction, border surveillance, maritime patrol across the Caribbean and Gulf. The aircraft was being pitched as a solar alternative to satellite coverage and as a longer-loitering complement to fast-jet ISR platforms like the MQ-4C Triton.

The Statement and the Silence

Piccard and Borschberg learned about the crash the same way most people did: through social media. The Solar Impulse team issued a short statement to Popular Science describing themselves as saddened by the loss of an important technological flagship.

Skydweller, by contrast, hasn’t responded publicly. That silence matters. Solar Impulse 2 was supposed to eventually return to Switzerland for permanent display at the Swiss Museum of Transport in Lucerne, per the original 2019 sale contract. That part of the deal is now unfulfillable.

The NTSB report points to a loss of power as the immediate cause of the crash. The deeper question of what failed inside the autonomy stack, the energy management software, or the solar-to-battery balance hasn’t been answered yet.

What the Loss Actually Costs

The hardware is one thing. The data is another. Solar Impulse 2 wasn’t just a museum piece in waiting. It was the test bed Skydweller had been building its production aircraft around, the proven airframe that validated everything from aeroelastic behavior at altitude to multi-day energy balance modeling. Losing it doesn’t end the program, but it resets a major piece of the validation work.

It also raises uncomfortable questions about the operational readiness of solar drones at this scale. The promise of 90-day flights has been the headline number Skydweller and similar programs have leaned on. The reality is that the longest demonstrated flight to date has been measured in days, not months. A power-loss crash in benign Gulf weather isn’t the failure mode investors and military procurement officers want to see.

DroneXL’s Take

Here’s what I find genuinely significant about this story. The crash isn’t just the loss of a famous aircraft. It’s a reminder that the gap between demonstration and operational reliability in the high-altitude pseudo-satellite category is still enormous, no matter how many press releases say otherwise.

BAE’s PHASA-35 has had its own struggles. Airbus shelved the Zephyr program after a string of crashes. Now Skydweller has lost its flagship airframe during what should have been a routine autonomous test. There’s a pattern, and the pattern is that flying a featherweight aircraft on borrowed sunlight for weeks at a time is harder than the renderings suggest.

The military case for these aircraft is real. A solar drone that can stay up for 30 days over the Caribbean costs less than launching another satellite, and the persistence-versus-revisit-rate math favors loitering aircraft for a lot of missions. The commercial case for telecom relay and disaster response is real too.

The engineering case is the part that keeps getting tested and keeps coming up short. Energy balance during night cycles, aeroelastic stability in turbulence, battery degradation over months of continuous use, autonomy stacks robust enough that nobody has to fish your aircraft out of the Gulf. None of these are solved problems. They’re hard problems, and the industry has been quietly losing aircraft while it works on them.

Solar Impulse 2 deserved a better ending than a Gulf of Mexico crash report. It was supposed to fly back to Lucerne. Instead, the airframe that proved sunlight could carry a human around the world is now wreckage, lost while pretending to be something it was never originally built to be.

Photo credit: Solar Impulse 2