For ecology and land-based field research

Field research, recorded where it happens.

ResearchOS runs your survey protocols and GPS-tracked field sessions, with iOS and Android apps that keep working out of range. Every record lands on an instance your group owns.

Tell us about your research See how it works

Distance: 3.4 km
Duration: 2 h 16 m
Records: 3 linked

Activity record

Vegetation quadrat

Site

Ridge plot 3

select

Quadrat size

1 m × 1 m

select

Dominant species

Kangaroo grass Themeda triandra

species

Native cover

42 %

number · %

Bare ground

11 %

number · %

−35.2809, 149.1300 ± 4 m

Saved offline

Fig. 1A patrol on North ridge with three linked quadrat records. Record 2 is open on the right. Illustrative example.

Your protocol, as a form the whole team fills the same way.

Coordinators define each survey protocol once as an ordered list of typed fields. Researchers capture records against it in the field, and the server checks every value against the template before it is stored.

Thirteen field types: Text, numbers with units and limits, counts, dates, yes/no, pick lists, species with scientific names, photos and point locations.
Context on every record: Time, coordinates, photos, notes and weather sit alongside your protocol values, so the record stands on its own.
Monitoring series: Key a template on a site or plot field and pick a numeric measure. Repeat visits line up as a timeline.
Validated at the source: Required fields, allowed options and min/max ranges are enforced, so bad values are caught while you are still on site.

Vegetation quadrat

6 fields

1Sitesite_idselect
2Quadrat sizequadrat_sizeselect
3Dominant speciesdominant_spspecies
4Native covernative_covernumber · %
5Bare groundbare_groundnumber · %
6Photo pointsphoto_pointsphoto

Monitoring series: site_id → native_cover

Fig. 2A quadrat template: six typed fields in order. Illustrative example.

Ridge plot 3 · native cover %

5 visits

Fig. 3Five visits to one plot as a monitoring series. Illustrative example.

Every survey walk keeps its track.

A patrol is a GPS-tracked field session: a transect run, a spotlighting walk, a camera-trap round. Start it, walk, and the app records track points as you go.

Linked records: Activity records captured during a patrol are tied to it and placed along the route.
Study-area zones: Draw zones as polygons and tag each patrol to the area it covered.
Distance and notes: Finish with a computed distance, your notes, and a full track you can review on the web.

time	lat	lon	acc m	alt m
07:42:10	−35.28314	149.12688	3.8	612
07:42:25	−35.28301	149.12711	4.1	614
07:42:40	−35.28287	149.12735	3.6	617
07:42:55	−35.28270	149.12752	4.4	619

Fig. 4Track points as the app stores them: time, position, accuracy and altitude. Illustrative example.

Connect to your ResearchOS instance

Instance address

field.mylab.example

→ https://field.mylab.example/api/v1/instance

Instance found

Example Ecology Group · API v1

Continue to sign in

Fig. 5First launch on iOS or Android. Illustrative example.

Built for the paddock, the gully and the ridge line.

Works with no signal: Records, photos and track points are stored on the device and sync when you are back in range. Each carries a device ID, so a retried upload never creates a duplicate.
GPS tracks with accuracy: Track points keep time, position, accuracy and altitude. A patrol captured entirely offline uploads in one pass.
Point the app at your instance: On first launch, enter your instance address. The app checks it, shows your organisation, then asks you to sign in. Admins can share a QR code from the web app to fill it in for the team.

One instance per research group, on a host you choose.

ResearchOS is not a shared cloud. Each deployment is a self-contained web app and Postgres database with its own sign-in and its own storage for photos. Nothing phones home to us.

Self-host with Docker

Run the container image next to Postgres on your own server or cloud account. Set the address it answers on, and optionally a sub-path if it sits behind an existing site.

APP_BASE_URL=https://researchos.mylab.example
BASE_PATH=/researchos      # optional
UPLOAD_DIR=/data/uploads
ALLOW_SIGNUP=false

Your accounts, your people

Sign-in lives on your instance too. There is no outside identity provider, and your group decides who gets in and what they can do.

No external sign-in service. Email and password accounts on your instance.
The first person to sign up becomes the admin. Everyone after joins by invite.
Roles for admins, coordinators, researchers and read-only viewers.

Protocols you can build on day one.

Examples of the survey templates groups set up in ResearchOS. Each one is just a list of fields, so you can match the method your project already uses.

Vegetation quadrat
Plot-based cover and composition
- site_id: select
- dominant_sp: species
- native_cover: number %
- photo_points: photo
Line-intercept transect
Cover along a measured line
- transect_id: select
- length_m: number m
- intercepts: count
- substrate: multiselect
Bird point count
Timed counts at fixed points
- point_id: select
- species: species
- count: count
- detection: select
Camera-trap check
Service visits and card swaps
- camera_id: select
- battery_ok: boolean
- card_swapped: boolean
- images: integer
Water quality sample
Field readings at a sampling site
- site_id: select
- ph: number
- turbidity: number NTU
- sampled_at: datetime
Soil core
Cores for lab analysis
- core_id: text
- depth_cm: number cm
- moisture: select
- location: location

Questions from the field.

Can I use ResearchOS now?

We are working with a small number of research groups first. Use the form at the bottom of this page to tell us about your fieldwork and the protocols you run, and we will be in touch.

Where is our data stored?

On your instance. Records, tracks and users live in the instance's Postgres database, and photos on its own storage volume. There is no shared ResearchOS cloud holding a copy.

Does the mobile app work without reception?

Yes. Templates are cached on the device, and records, photos and GPS tracks are queued locally until you have a connection again.

Can we define our own survey protocols?

Yes. Admins and coordinators build activity templates in the web app from the available field types, including units, limits, pick lists and required fields.

Can several labs share one instance?

Each instance serves one organisation. Groups that want separate data run separate instances, each on its own address.

What does it cost?

We are working through pricing with early-access groups. Talk to us about your setup.

Tell us about your research.

What you survey, where, and how you record it today. It helps us understand whether ResearchOS fits your fieldwork, and what to build next.

Prefer email? Write to hello@yumait.au.

Your nameEmailOrganisationType of organisationYour roleRegion or country (optional)

What research do you do, or want to do with ResearchOS?Study sites, species or habitats, how often you survey, what you use today.

Team size (optional)Hosting (optional)

Anything else (optional)

Field research, recorded where it happens.

Your protocol, as a form the whole team fills the same way.

Every survey walk keeps its track.

Built for the paddock, the gully and the ridge line.

One instance per research group, on a host you choose.

Self-host with Docker

Your accounts, your people

Protocols you can build on day one.

Vegetation quadrat

Line-intercept transect

Bird point count

Camera-trap check

Water quality sample

Soil core

Questions from the field.

Tell us about your research.