Fits the clock, not the way back
A stop fits the EVA time budget but lies beyond the limit for walking back to the habitat.
A candidate route across real NASA Mars data, with a source for every number.
ARES brings NASA Mars data into one local view and proposes a candidate route between science stops. Time, climb and the limit for walking back sit next to it, and every number says where it came from. It is our entry in the NASA Space Apps Challenge 2026, Interplanetary Survival Guide: Martian Map.
A stop fits the EVA time budget but lies beyond the limit for walking back to the habitat.
A slope looks mild on a 463 m grid and turns steep on a 1 m terrain model.
Misaligned layers create false hazards, or false “safe” corridors.
The question ARES is built around: can the crew walk back from every point of the route? Existing public tools are made for browsing regions. ARES is made for one walk.
A one-minute walkthrough: place stops, watch the route update as you move them, compare an alternative, and print the route card.
Status is shown on every card. The planner runs on sample data today; real terrain data is the next step.
Click to set a start and stops, drag to move them. The route recomputes after each edit and the old one fades until the new one arrives.
Route segments are coded by slope, and hazard zones are hatched with a label. Colour is never the only signal: every status has an icon and text.
Distance against elevation, coloured by slope class, with stop markers. Hover it and the position shows on the map, and the other way round.
Overlay a second route and read the difference in length, steepest slope and time.
A one-page A5 card with a route sketch, stop list, slope table by leg and the assumptions behind it.
Compass bar, next stop, slope and nearest hazard. It is a mockup: position comes from a slider, not a live feed.
Place up to six science targets. Each has an objective, its evidence, a confidence level and a rover image with credit.
Orbital imagery, terrain model, slope and rover images in one window, with resolution and coverage shown for each.
An A* search on a 1 m grid from the habitat through your stops and back. If no route exists, ARES names the blocking constraint instead of drawing a straight line.
Each leg gets its 3-D distance, climb, time and a relative effort index, checked against the EVA time budget and the walk-back limit.
Unknown terrain is penalised, never treated as free to cross. Steps 3 and 4 are in progress; they are not in the prototype yet.
Each layer shows its mission, instrument, product, resolution and coverage, marked native, coarse or missing. A gap is never treated as safe ground.
| Layer | Source | Resolution | Role |
|---|---|---|---|
| Orthoimage | HiRISE | 25 cm | Visual base |
| Elevation | HiRISE terrain model (DTM) | 1 m | Slope, roughness, routing |
| Slope and roughness | Derived from the terrain model | 1 m | Hazard mask |
| Thermal inertia | THEMIS | 100 m | Context only, never a per-cell hazard |
| Rover images | Curiosity Mastcam and Navcam | – | First-person view per stop, with credit |
| Radiation | Curiosity RAD | point | Card only, not a map |
| Weather | Curiosity REMS | point | Historical climate card |
The area of interest is the boxwork on Mount Sharp in Gale crater (search frame 137–138° E, 5.5–4.3° S). Curiosity has surface radiation and weather records there, thousands of rover images and detailed orbital imagery. Boxwork ridges are 5–9 m wide and 1–2 m high, so the best path is not a straight line.
Each number carries a value, a unit, a source and one of four labels. These are the starting values behind a route; planners can change them and the route is rebuilt.
| Parameter | Default | Label | Source |
|---|---|---|---|
| Walking speed | 2 km/h | assumption | Lunar ConOps EVA-EXP-0042 Rev B. 4 km/h only as a sensitivity case. |
| Hard slope limit | 15° (20° option) | proxy | Lunar analog, Peña-Asensio 2024. Slopes on a 1 m grid look steeper than on 5 m. |
| Walk-back limit | 1 h or 3.5 km | assumption | ICES-2026-127. The lunar 10 km is kept only for comparison. |
| Longest EVA | 8 h (4 h option) | assumption | Reference architecture, Rucker and Hoffman 2020 |
| Mars gravity | 3.73 m/s² | established | NASA NSSDC Mars fact sheet |
| Cosmic-ray dose at Gale | 0.64 ± 0.12 mSv/day | measured | Hassler et al. 2014, Science, Table 1 |
SEXTANT (MIT) already did metabolic A* routing and a best path home. We do not present that as new. What ARES adds is data from several Mars missions on one site, a walk-back limited by consumables with a point-of-no-return marker, and a source label on every number.
AI/ML, Frontend, Captain
Backend
Research, Validation
Visuals, AI generation
We worked on everything together and complemented each other.
Open the map, place a few stops and see a candidate route with its slope, climb and time.
Launch the app