Skip to main content

Real-time observability and safety for robotics simulations across all platforms

Project description

Oculus SDK - Research & Implementation

Modular, scalable SDK for robotics observability and safety.

Installation

# Basic (all platforms)
pip install oculus-robotics

# With platform-specific extras
pip install oculus-robotics[isaac-lab]
pip install oculus-robotics[ros2]
pip install oculus-robotics[mujoco]

Quick Start

from oculus.platforms.isaac_lab import IsaacLabTracer

with IsaacLabTracer(project="my_simulation") as tracer:
    for step in range(1000):
        # Your simulation code
        state = get_robot_state()
        tracer.auto_capture(state)

Development Setup

# Clone and install in dev mode
cd oculus-sdk
pip install -e ".[dev]"

# Run tests
pytest tests/

Package Structure

oculus/
├── core/              # Base tracer, connection, auth
├── platforms/         # Isaac Sim, Lab, ROS2, Mujoco, etc.
├── safety/            # Fall prevention, collision avoidance
├── prediction/        # Fall prediction, anomaly detection
└── research/          # Research paper implementations

Platform-Specific Usage

Isaac Sim

from oculus.platforms.isaac_sim import IsaacSimTracer

Isaac Lab

from oculus.platforms.isaac_lab import IsaacLabTracer

ROS 2

from oculus.platforms.ros2 import ROS2Tracer

Mujoco

from oculus.platforms.mujoco import MujocoTracer

Safety Algorithms

from oculus.safety.fall_prevention import QuadrupedSafeFall
from oculus.prediction import FallPredictor

safety = QuadrupedSafeFall(robot_type="unitree_go1")
predictor = FallPredictor()

result = safety.check_fall_risk(robot_state)
if result.intervention_needed:
    correction = safety.prevent_fall(robot_state, result)

Research Implementations

See oculus/research/ for implementations of:

  • Radium paper (Unitree safe fall)
  • Custom algorithms from latest research

Contributing

  1. Study research paper
  2. Implement in oculus/research/new_algorithm/
  3. Test in oculus-testing/
  4. Submit PR with paper reference

Publishing

python -m build
twine upload dist/*

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

oculus_robotics-1.0.0.tar.gz (38.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

oculus_robotics-1.0.0-py3-none-any.whl (48.7 kB view details)

Uploaded Python 3

File details

Details for the file oculus_robotics-1.0.0.tar.gz.

File metadata

  • Download URL: oculus_robotics-1.0.0.tar.gz
  • Upload date:
  • Size: 38.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.9.13

File hashes

Hashes for oculus_robotics-1.0.0.tar.gz
Algorithm Hash digest
SHA256 8f8c20861eeb2d2a745feb77832e90f0fcc735dee09dd0ed4814aa9a6cd3bc26
MD5 61f091fe7846ccf88cf6bd38025deb42
BLAKE2b-256 65f75fafaffa25d8c1fef38fc983c31a6afab09fa911508475672cbf8aee02dc

See more details on using hashes here.

File details

Details for the file oculus_robotics-1.0.0-py3-none-any.whl.

File metadata

File hashes

Hashes for oculus_robotics-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 3b2e47751845f1647212f9a53c614b4c0fb24f4958e5f242dc6acce9a48a4ab6
MD5 7cd6d5976bb82ac6a6277780924623dc
BLAKE2b-256 c8399e010d6229f23a1319dceeb28fec337014cc0198ad153ccc8dfbc89a5e4e

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page