Skip to content

BulletLab v0.2.0

Choose a tag to compare

View all tags
@NuclearVenom NuclearVenom released this 01 Jul 12:03
· 19 commits to main since this release
v0.2.0
dd43d8c

This release introduces the first official integration between BulletLab and BulletLab Arsenal, making verified robotics assets installable and directly loadable from the official registry while significantly improving the project's architecture, documentation, and overall developer experience.

Highlights

BulletLab Arsenal Integration

  • Introduced Robot.install() for permanently installing verified robot packages from BulletLab Arsenal
  • Extended Robot.load() to support direct loading from Arsenal using the new URI format:
Robot.load("arsenal:reference_bot")
Robot.load("arsenal:unitree_g1/g1_29dof")
  • Added automatic temporary caching for Arsenal-loaded assets with session-based cleanup
  • Preserved full backward compatibility with existing local file loading

Package Management

  • Added support for loading default robot models or explicitly selecting models within a package
  • Implemented automatic dependency resolution for URDFs and required mesh assets
  • Added configurable installation locations for permanent package downloads
  • Designed the infrastructure for future support of Worlds, Sensors, Controllers, Datasets and Benchmarks

Documentation

  • Introduced comprehensive documentation for BulletLab Arsenal and its role within the BulletLab ecosystem
  • Updated the README to present Arsenal as the official package registry for BulletLab
  • Added installation and loading examples covering both local and Arsenal-based workflows
  • Improved project navigation, examples, API documentation, and user guides

Project Improvements

  • Performed a repository-wide audit of the codebase
  • Removed obsolete code, outdated documentation, and inconsistencies
  • Improved API consistency and internal architecture
  • Refined error handling and overall code quality
  • Polished the project for a more professional open-source release

About BulletLab

BulletLab is a robotics experimentation framework built on PyBullet that transforms robots into intuitive Python objects while providing modern tooling for simulation, visualization, telemetry, and robotics experimentation.

Its official package registry, BulletLab Arsenal, provides curated and verified robotics assets—including robots, worlds, sensors, controllers, datasets, and benchmarks—that can be installed or loaded directly from within BulletLab.

Installation

pip install --upgrade bulletlab

Links

Documentation: https://nuclearvenom.github.io/BulletLab/
Repository: https://github.com/NuclearVenom/BulletLab
BulletLab Arsenal: https://github.com/NuclearVenom/BulletLab-Arsenal

Assets 2
Loading