RVDLAB

Open infrastructure for rendezvous & docking.

Simulation, control and reproducible experiments for spacecraft research.
Configure a run, connect your controller, and keep the data needed to replay it.

What is RVDLAB?

RVDLAB is the terminal interface and experimental workflow for RVD-RailBench. It brings simulation, control, configuration and records into one research stack. Read the docs.

Configurable simulation

Choose ideal or bounded simulation, initial conditions, noise and delay.

Controller interfaces

Keep algorithm implementations separate from the experiment runtime.

Versioned conditions

Record the build, model, configuration and random seed.

Complete records

Save telemetry, controller inputs, events and the final result.

Controller replay

Check integrity and compare controller behavior against recorded inputs.

Ground testbed

A research platform for rail motion, sensors and target motion. Hardware integration is in progress.

See what
RVDLAB can do

Configure an experiment, run it, inspect the record, and replay the controller from one terminal workflow.

Run experiments from your terminal

Check your environment, select a configuration and launch a simulation. RVDLAB brings the C++ runtime and experiment tools into a terminal interface.

Get started
RVDLAB v0.4.1 home screen with simulation, policy and device status
RVDLAB v0.4.1 · Linux x64 · Simulation only Enlarge ↗

Keep the conditions with the experiment

Apply configuration files in order, then save the resolved settings with the run. Keep the model, build version, seed and recorded inputs alongside the trajectory.

Explore experiment records
experiments/example/

Run provenance

metadata.json

Model and artifact identifiers, SHA-256, runtime backend, compiler, build commit, seed and clock mode.

Replay the recorded controller inputs

Check record integrity, restore the initial state and replay the controller. Compare actions, next states and flags using the high-precision record.

Read about replay
Replay an experiment
build/release/rail_replay experiments/example
Inputs
Recorded observations
Comparison
Actions, state and flags
Hardware commands
None

Get started

Build from source, check the runtime, and run the documented simulation example.

Install

Set up Python and the C++ runtime from a fixed source version.

Installation

Run

Run a 12-step example and save it to a new experiment directory.

First experiment

Replay

Check the record and reproduce the controller calculation.

Replay guide

Build with us

Contribute a controller, a device adapter, a reproducible example or a documentation improvement.

View the source ↗Open an issue ↗