Fukushima Robot Test Field lunar analog ('Moon/Mars Garden')
Unreviewed draft
This page was built from a teammate-supplied list. The text is teammate-written, unreviewed. Sources pending.
Summary
not provided
Location
- Country
- Japan (Fukushima)
- Coordinates
- 37.63, 141.00 (approximate)
- Elevation (m)
- 10
- Year first used as an analog
- 2025
Terrain and geology
not provided
Climate
not provided
Parameters matched
not provided
Why it is an analog
Newest Asian lunar test field, with engineered terrain and robotics infrastructure.
Key differences: where the analogy breaks down
Small artificial bed on a coastal, humid, built-up site.
Experiments done
University of Aizu Hakoniwa project (2025) building lunar analog terrain inside the national Fukushima Robot Test Field for rover and construction-robot tests.
Experiments proposed
Rover autonomy on engineered lunar terrain; lunar construction robotics (excavation, berm building); teleoperation from a remote control room.
How to adjust it
Already artificial: add fine simulant and polar-light rigs.
Access
not provided
Hazards
not provided
Sources
Sources pending (no sources in CSV).
v1 score, percentile and overlap flag
A descriptive lookup of scores v1 already stored. It is not a test, and a high percentile here does not show that this is a good analog.
v1 used this site
No. The nearest v1 site, "samoylov", is about 3947.8 km away.
| Score view | Result | Percentile | Score | Distance to tile (km) | Overlap flag |
|---|---|---|---|---|---|
| Connecting Ridge C1, Base | Missing datano tile in range passed; some tiles lack data | not provided | not provided | not provided | not provided |
| Connecting Ridge C1, Landing and mobility | Missing datano tile in range passed; some tiles lack data | not provided | not provided | not provided | not provided |
| Connecting Ridge C1, Power and night | Missing datano tile in range passed; some tiles lack data | not provided | not provided | not provided | not provided |
| Shackleton rim, Base | Missing datano tile in range passed; some tiles lack data | not provided | not provided | not provided | not provided |
| Shackleton rim, Landing and mobility | Missing datano tile in range passed; some tiles lack data | not provided | not provided | not provided | not provided |
| Shackleton rim, Power and night | Missing datano tile in range passed; some tiles lack data | not provided | not provided | not provided | not provided |
Overlap flag: whether the tile used for the score is one v1 flagged as stable in its top group across the Monte Carlo runs.
What these scores can and cannot tell you
- This is a lookup of scores v1 already stored. It is descriptive and is not a test of the method.
- A site is matched to the best-scoring tile that passed the gates within the match radius. A different radius would give different numbers.
- Tiles are large compared with most sites, so a score describes the surrounding area, not the site itself.
- Scores exist only for tiles that passed the land-cover and data gates. A gated-out or unscored site has no score, which says nothing about how good an analog it is.
- Percentile is the share of scored tiles in that view with a lower score.
- The Moon score is effectively terrain only. Mars is a low-confidence secondary result.
- v1 failed its negative controls, so a high percentile here does not show that a site is a good analog.
- The ranking is not checked against independent data. Gravity, air, dust and radiation are not scored.
- Sites are compared with views of their own body only.
- Match radius: 20 km, the radius v1 used.