User Guide

Welcome to LARUN.SPACE! This guide will help you discover exoplanets using our AI-powered detection system. No astronomy degree required.

What is LARUN?

LARUN (Light Analysis Research Utility Network) is a federation of TinyML models designed to detect exoplanets in NASA telescope data. Each model specializes in a different aspect of astronomical analysis, from transit detection to stellar classification.

First Steps

1. Create an Account (Optional)

Free users can analyze up to 10 targets per month. Create an account to:

2. Navigate to the App

Click Get Started to open the chat interface. This is where all the magic happens.

3. Choose Your Target

You can search for any star observed by NASA's TESS or Kepler missions. Common identifiers:

Chat Interface

LARUN uses a conversational interface. Simply type what you want to do in natural language.

Example Commands

# Search for transit signals
"Search for transits in TIC 307210830"
"Analyze TIC 12345678 for planets"
"Look for periodic dips in TOI-700"

# Get stellar information
"What type of star is TIC 307210830?"
"Tell me about the host star"
"Is this a red dwarf?"

# Habitability analysis
"Is this planet in the habitable zone?"
"Could this planet have liquid water?"
"Calculate equilibrium temperature"

# Generate reports
"Create a report for this candidate"
"Generate PDF summary"
"Export my results"

Understanding the Response

When you analyze a target, LARUN will:

  1. Fetch Data: Download light curve from NASA archives
  2. Preprocess: Clean the data, remove outliers, detrend
  3. Detect: Run TinyML models to find transit signals
  4. Analyze: Fit parameters, estimate planet properties
  5. Report: Present results in an easy-to-understand format

Understanding Results

Key Parameters

Parameter Description Example
Period Time for one complete orbit (days) 3.42 days
Depth How much the star dims during transit (%) 0.23%
Duration How long the transit lasts (hours) 2.5 hours
SNR Signal-to-noise ratio (higher is better) 12.4
R_p Planet radius (Earth radii) 1.8 R_Earth
FPP False Positive Probability (lower is better) 0.02

Disposition Categories

VALIDATED

FPP < 1%. Very likely a real planet. Suitable for follow-up.

CANDIDATE

FPP 1-50%. Promising signal that needs more analysis.

LIKELY_FP

FPP > 50%. Probably an eclipsing binary or other false positive.

The most common workflow is searching for transit signals in a star's light curve.

1

Enter Target ID

Type: "Search for transits in TIC 307210830"

2

View Light Curve

LARUN displays the raw and detrended light curve

3

BLS Analysis

Box Least Squares detects periodic dips

4

TinyML Detection

Neural network confirms transit signal

5

Parameter Fitting

Transit model fitted to derive planet properties

Candidate Vetting

Not every transit signal is a real planet. LARUN helps you vet candidates:

False Positive Scenarios

Vetting Commands

"Calculate false positive probability"
"Check for secondary eclipse"
"Look for odd-even depth differences"
"Analyze centroid motion"

Habitability Analysis

Once you've found a candidate, LARUN can assess its potential habitability.

What We Calculate

"Is this planet in the habitable zone?"
"What is the equilibrium temperature?"
"Compare to Earth's habitability"

Report Generation

Generate publication-ready reports for your discoveries.

Report Contents

Export Formats

Using Custom Data

You can upload your own light curve data for analysis.

Supported Formats

CSV Format Example

time,flux,flux_err
2458765.0,1.00023,0.00012
2458765.1,0.99987,0.00011
2458765.2,0.99234,0.00013
...
Data Requirements
  • Time in BJD (Barycentric Julian Date) or relative days
  • Flux normalized to ~1.0
  • At least 1000 data points recommended
  • Cadence of 2 minutes or better for small planets

API Access

Pro and Scientist tier users get API access for programmatic analysis.

Getting Your API Key

  1. Sign in to your account
  2. Go to Settings > API Keys
  3. Click "Generate New Key"
  4. Copy and store securely

Example Request

curl -X POST https://api.larun.space/v1/pipeline \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tic_id": "307210830",
    "full_report": true
  }'

See the API Documentation for complete reference.

Batch Processing

Analyze multiple targets at once (Scientist tier only).

Batch Commands

"Analyze TIC 1, TIC 2, TIC 3"
"Run pipeline on TOI-700, TOI-1338, TOI-1452"
"Process all targets in my list"

Uploading a Target List

Upload a CSV file with one TIC ID per line:

tic_id
307210830
123456789
987654321

Ready to Start Discovering?

Jump into the app and find your first exoplanet candidate!

Launch LARUN App