Programming Robot Intelligence
Robotics applications are typically distributed, made up of a collection of concurrent asynchronous components which communicate using some middleware (ROS messages, ICE, DDS...). Building robotics applications is a complex task. Integrating existing nodes or libraries, which provide already solved functionality, and using several tools may increase the software robustness and shorten the development time. JdeRobot toolkit provides several tools, libraries and reusable nodes for Robotics and Computer Vision.
|ROS friendly (full compatible with ROS-Kinetic)|
|Easy installation from debian, PIP and npm packages|
Our (international) community mainly works on five development areas:
|Robot Programming Tools||VisualStates tool for robot programming with automata|
|WebSim2D robot simulator with web technologies|
|Scratch4Robots tool for visual robot programming|
|Education in Robotics and Computer Vision||Robotics-Academy for engineering students|
|Unibotics online framework for engineering students|
|SLAM, visual localization||SDslam visual SLAM algorithm|
|slam-testbed tool for evaluation of SLAM algorithms|
|DeepLearning, visual perception||DetectionSuite tool for evaluation of detection networks|
|Object Detector see deep learning networks in action|
|FPGAs in robotics||neuralFPGA running deeplearning networks on FPGAs|
|FPGA-robotics designing robot intelligence with IceStudio blocks|
|Drones, autonomous behaviors|
We are registered as non-profit organization in Spain Ref.#615800.
- JdeRobot was accepted as a mentoring organization for Google Summer of Code 2019, several students will be funded by Google to work in open projects of our organizaton. Check the proposed ideas list and our candidate selection process.
- The last stable release, JdeRobot/base 5.6.7 has been released (2019/03/12): full compatible with ROS Kinetic.
- 3rd edition of Program-A-Robot Challenge was celebrated inside the International Conference on Intelligent Robots and Systems (IROS 2018) (2018/10/03). Watch the summary video: a drone (cat) searches and pursues a moving target (mouse).
|SDSLAM node with Augmented Reality (more)||Drone exercise in Robotics Academy using MavROS and PX4 (more)|
The last stable release is 5.6.4
We are working on:
- Update of underlying infrastructure: jump to Ubuntu 18.04, ROS Melodic Morenia, Gazebo9 simulator, Python 3.5.
- Use of reconfigurable hardware (FPGAs) in robotics, with third party open tools like IceStudio
- VisualStates tool for visual programming of the robot intelligence with Finite State Machines. It creates a C++ or a Python component from the visual description of the automata.
- Scratch4Robots tool for programming of TurtleBot and drones using Scratch visual language
- Robotics Academy: a framework to learn robotics and computer vision with drones, autonomous cars.... It is a collection of Python programmed exercises. Include a web service for this.
Compatibility with ROS robotics middleware: interoperation between ROS nodes and JdeRobot components, use of ROS drivers, use of ROS bag files...JdeRobot (>= 5.6) is fully compatible with ROS Kinetic including its official debian packages
Update of underlying infrastructure: jump to Ubuntu 16.04, OpenCV-3, migration to Gazebo-7 simulator (revisit existing plugins and models), PCL-1.8, ICE-3.6
Use of Web technologies in robotics: ICE-JS, WebRTC, HTML5, electron, nodeJS...Many webtools have been developed: viewers, teleoperators and drivers.
JdeRobot is open source. All its code is licensed under GPL v3.
All the documentation of JdeRobot project is licensed under Creative Commons by-sa