You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As Dora continues to grow and various development activities take place, it's important to inform people about the overall structure and how they fit into the bigger picture. Here are some modules we might unify when communicating with those unfamiliar with Dora.
Feature
ROS2
ISAAC
DORA
Harix OS
Conclusion
Robot control
ROS2_control
Motion Generation Extension
NA
RCU
manipulator abstract interface
Robot Navigation
nav2
Navigation or nav2 through bridge
NA
RCU
mobile robot navigation abstract interface
AI
ROS-LLM
GEMS
NA
Harix AI
built-in Robot specific AI Library
Sim
Gazebo, Copperslim, Webots, carla
Isaac SIM
Carla, RViz through ROS2
Sim
wide robot model is needed
SDK
NA
SDK
Dora Core
RDK
Lib/Developer examples/docs
Flow UI
NA
ISAAC UI
NA
IDE
visualize and design complex systems
Since we have initiated DORA-ARM, I would like to expand on it to provide a broader perspective on how our future work fits into the larger vision.
Here is the proposed diagram:
The framework layer is an abstraction designed to simplify application development. We anticipate that DORA Control, Navigate, and AI will be our primary focus for future work. These packages can be installed separately like cargo install or pip install.
The Dora AI packages include some common libraries to support the applications such as hugging-face, open-mmlab etc.
The Dora control packages are libraries designed to solve this problem. It focuses on providing universal, simple and well-structured Python interfaces for robots, grippers and cameras, IMU, Lidar, etc. These interfaces provide implementations of these interfaces for general-purpose hardware. This makes the code portable between different robot setups.
Lidar
Camera
IMU
Odometry
Robot Manipulator Interface
moveJ
moveL abstraction on top of vendor driver support
TF
IK planner
Joint Path
Collision Avoidance
Eye-hand calibration
Aruco/April Tags
Sensor recording
The DORA applications serve as reference designs built on top of the DORA framework and core. For instance, DORA-ARM is a reference design based on ALOHA, addressing AI and manipulator usage. Similarly, DORA-AMR and DORA-Autoware are designed to address navigation.
One advantage is that developers of various skill levels have multiple ways to contribute. Some may prefer to work at the application level without needing in-depth expertise in DORA core. This flexibility makes it easier for developers to adopt and contribute to the platform.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
As Dora continues to grow and various development activities take place, it's important to inform people about the overall structure and how they fit into the bigger picture. Here are some modules we might unify when communicating with those unfamiliar with Dora.
Since we have initiated DORA-ARM, I would like to expand on it to provide a broader perspective on how our future work fits into the larger vision.
Here is the proposed diagram:
The framework layer is an abstraction designed to simplify application development. We anticipate that DORA Control, Navigate, and AI will be our primary focus for future work. These packages can be installed separately like cargo install or pip install.
The Dora AI packages include some common libraries to support the applications such as hugging-face, open-mmlab etc.
The Dora control packages are libraries designed to solve this problem. It focuses on providing universal, simple and well-structured Python interfaces for robots, grippers and cameras, IMU, Lidar, etc. These interfaces provide implementations of these interfaces for general-purpose hardware. This makes the code portable between different robot setups.
The DORA applications serve as reference designs built on top of the DORA framework and core. For instance, DORA-ARM is a reference design based on ALOHA, addressing AI and manipulator usage. Similarly, DORA-AMR and DORA-Autoware are designed to address navigation.
One advantage is that developers of various skill levels have multiple ways to contribute. Some may prefer to work at the application level without needing in-depth expertise in DORA core. This flexibility makes it easier for developers to adopt and contribute to the platform.
Beta Was this translation helpful? Give feedback.
All reactions