SoftBank Demonstrates First Ground‑Laser Tracking of a Stratospheric HAPS

In late August 2026, a modest‑sized laser sitting on a tabletop in Japan managed to “see” a flying platform cruising roughly 20 km above the ocean. The trick? A simple corner‑cube reflector glued to the underside of a high‑altitude platform (HAPS) owned by U.S.‑based Sceye. When the laser pulse hit the reflector, it bounced straight back to its source, letting the ground team keep tabs on a moving target far higher than any traditional drone.

At first glance the experiment might look like a kid’s science‑fair project – just a laser, a mirror, and a balloon‑like aircraft. In reality it was a carefully choreographed test of what engineers call pointing, acquisition and tracking (PAT). Those three steps are the nervous system of any optical communication link: you have to locate the aircraft, lock a razor‑thin beam onto it, and then stay locked as it drifts, rolls and pitches in the wind.

The reflector used is a corner‑cube reflector (CCR). Its geometry sends incoming light back toward where it came from, no matter the angle of arrival. By mounting a CCR on the HAPS, SoftBank and its partners could focus on the hard part – keeping the laser aimed at a target that’s constantly shifting in three‑dimensional space – without worrying about transmitting actual data.

During the two‑day trial, held over the waters off Cape Muroto in Kochi Prefecture, the portable laser‑ranging unit on the ground continuously fired short pulses upward. Each pulse that made the round‑trip was recorded, giving the team a stream of timing data that revealed not just the platform’s position but also how much the atmosphere dimmed the beam. That kind of attenuation data is gold for designing future HAPS‑to‑ground optical terminals, because the thinner, faster laser you need, the more you have to know about how clouds, humidity and aerosols will eat away at the signal.

It’s worth noting that the laser power was deliberately kept low – nowhere near the levels that could pose a hazard to aircraft. In fact, the system was programmed to shut off instantly if a manned plane crossed the beam’s path. Safety first, always.

This isn’t the first time anyone has tried to talk to the sky with light. Back in 2005, Germany’s DLR launched STROPEX, beaming a 1.25 Gbit/s downlink from a 22 km‑high balloon. The U.S. Navy’s SOLD project has also been busy with stratospheric optical links. What sets SoftBank’s demo apart is the pure round‑trip geometry: ground → HAPS → ground, using a passive reflector rather than an active transceiver. It’s a clean way to validate the PAT algorithms before adding the complexity of a full‑blown optical modem.

Looking ahead, SoftBank says the next step is to replace the CCR with an actual optical transceiver that can both send and receive data. Because the tracking software used for the ranging test is already built into the communications prototype, the jump from “we can see you” to “we can talk to you” should be smoother than starting from scratch.

The ultimate vision, according to SoftBank, is a three‑dimensional mesh: ground stations talking to HAPS aircraft, which in turn hand off data to low‑Earth‑orbit (LEO) satellites. In such a network, laser links would provide the high‑capacity backbone, while radio waves take care of the last‑mile connections to phones and other devices. SoftBank is already trialing smartphone‑grade services from Sceye’s HAPS, and hopes to roll out commercial offerings by 2027.

In short, the experiment proved a fundamental piece of the puzzle – you can reliably find, lock onto, and keep a laser beam glued to a platform twenty kilometres up, even when the atmosphere is throwing a few curveballs your way. That’s a big deal for anyone dreaming of a space‑and‑sky internet that doesn’t rely solely on rockets.