Isaac Perceptor

Isaac Perceptor is a camera-based 3D perception system for mobile robots. It provides robust odometry together with local 3D scene reconstruction to enable autonomous navigation applications.

Isaac Perceptor leverages multiple other Isaac ROS modules:

• Isaac ROS Nova for time-synchronized multi-cam data.

• Isaac ROS Visual SLAM for GPU-accelerated camera-based odometry.

• Isaac ROS Depth Estimation for learning-based stereo-depth estimation.

• Isaac ROS Nvblox for GPU-accelerated local 3D scene reconstruction.

• Isaac ROS Image Pipeline for GPU-accelerated image processing.

For details on how these packages are combined in Isaac Perceptor see:

• Isaac Perceptor Technical Details

Tutorials

Isaac Perceptor supports running on the following robot platforms and simulation environment,

• robots equipped with the Nova Orin Developer Kit

• Nova Carter

• Isaac Sim

Running Isaac Perceptor on the Nova Orin Developer Kit could serve as a quickstart.

On the Nova Orin Developer Kit

Review the following guide to run camera-based 3D perception with Isaac Perceptor on the Nova Orin Developer Kit, which can be mounted on your robot.

• Camera-based Perception with Isaac Perceptor on a Robot with the Nova Orin Developer Kit

On the Nova Carter

Review the following guide to run camera-based 3D perception, and autonomous navigation with Isaac Perceptor on the Nova Carter.

• Camera-based Perception with Isaac Perceptor on the Nova Carter

In Isaac Sim

Review the following tutorial to run camera-based 3D perception, and autonomous navigation with Isaac Perceptor in Isaac Sim.

• Camera-based Perception with Isaac Perceptor in Isaac Sim

Repositories and Packages

The Isaac ROS implementation of Isaac Perceptor is in the following repository:

• Isaac Perceptor

Troubleshooting

• Troubleshooting