Google Summer of Code 2018 Ideas
26 April 2018
Here we have a list of the final ideas and students selected for GSOC’ 2018:
Sayali Deshpande
Idea: Emotion Recognition Component for Learnbot
Mentors: Iván Barbecho Delgado and Pilar Bachiller
Mayank Kumar
Idea: Extending RCManager functionalities
Mentors: Esteban Martinena and Pablo Bustos
Akash Kumar Singh
Idea: Gazebo-RoboComp Integration
Mentors: Marco A Gutiérrez and Ramon Cintas
Maria Bohórquez
Idea: Generation of new use cases, tutorials and reference information for RoboComp
Mentors: Marco A Gutiérrez
Araceli Vega Magro
Idea: Improving the Human-centered Robot Navigation Agent
Mentors: Luis Jesús Manso and Pedro Núñez Trujillo
Piyush Pilaniya
Idea: Integrating MORSE Human-robot interaction simulator with RoboComp
Mentors: Luis Jesús Manso and Pablo Bustos
José Alberto Andújar
Idea: JavaScript support
Mentors: Nicolas Gonzalez Flores and Luis V. Calderita
Prinshu Kumar
Idea: Learnblock (Learnbot programming language) extension for collaborative robotics
Mentors: Marco A Gutiérrez and Iván Barbecho Delgado
Sparsh Garg
Idea: RCIS: improving RoboComp simulator with contact physics
Mentors: Pablo Bustos and Ramon Cintas
Cristina Mendoza Gutiérrez
Idea: RCIS: improving RoboComp simulator with contact physics
Mentors: Pablo Bustos and Ramon Cintas
Google Summer of Code 2018 Ideas
23 Jan 2018
General information on applications
This is the list of ideas for the Google Summer of Code 2018, if you are interested in any of the ideas listed below or you think you can propose something interesting to improve RoboComp we encourage you to apply.
- It is important to first familiarize with the software (https://github.com/robocomp/robocomp).
- You can go through the available tutorials and direct your questions to gitter chat (link in contact section).
- Please read all the information posted in this page before applying.
- Make sure you are familiar with the required skills for the idea.
- Since several of the mentioned RoboComp tools and components are not explained here to keep this list short, we encourage everyone to check the RoboComp documentation linked below.
- Mentors and backup mentors are listed right after the idea explanation. All mentors contact info is at the end of the page. Contact them directly for specific questions on the idea.
Robocomp installation and tutorials: https://robocomp.readthedocs.io/en/highlyunstable/
Where can I start and what to include on my application?
You are encouraged to go through these steps for a better understanding and follow-up of your application:
- Download and install RoboComp: https://github.com/robocomp/robocomp/blob/master/README.md.
- Follow the tutorials: https://github.com/robocomp/robocomp/blob/master/doc/README.md.
- Once you are familiar with RoboComp and the components and tools involved in the particular idea you want to contribute to, try to understand how these components/tools work and, if possible, their design.
- Participate in gitter asking any question you might have.
- Create a schedule with the milestones you plan to follow during the GSoC 2017 program.
- Send the schedule and any code you might have written along with your CV and other application materials to the mentor of your idea and the backup mentor.
For general questions about RoboComp please use: The gitter chat.
IDEAS LIST for GSoC 2018
1. RCManager extended functionalities: a tool to deploy and monitor large sets of components.
Currently RoboComp has a tool to deploy components that is being improved through several GSoC editions. It’s name is RCManager and is used on a daily basis by all the people that use RoboComp to program robots. Since this tool is crucial to the software development process and since robot software is all the time increasing its complexity as a large-scale distributed system, we need to improve this tool as much as we can.
- The first extension is to make the tool access remote RoboComp installations to create a list of potential components to be added to the deployment set.
- The second extension tackles the need to group sets of components into higher order entities, so visualization is simplified.
- A third extension is to include the capability to probe the edges in the graph of components so a pop-up window would show in real time the traffic moving through it. When connections between components use publish/subscribe modality the probing is easy, but when communication is done using the pull/request modality things get more complicated.
Where to start and what to include in my application?
- Run a test with the RCManager tool and understand how it works
- Have a look at the tool’s code and check for how to make modifications so a deployment file can be automatically generated in a painless way through the tool
- Provide some example of a small modification or improvement in the tool.
Technologies involved: C++, Python, Qt, OPenSceneGraph
Mentors: Esteban Martirena, Pablo Bustos
Backup Mentor: Luis J. Manso
2. RCIS: improving RoboComp simulator with contact physics
RCIS is RoboComp’s robot simulator and a very valuable tool that has been developed for the last years to suit our needs. However, as robots get more and more complex more advanced functions are needed in RCIS. This project’s goal is the extension of RCIS with a new set of functionalities dealing with contact physics. Without incurring in a full fledged physics engine in this work, the student will use the existing FCL collision detection library already included in RCIS to program different ways to control the behavior of simulated objects, robots and humans. The skill needed for this project include C++, OpenSceneGraph and knowledge of basic physics.
Technologies involved: C++, Python, Qt, OPenSceneGraph
Mentors: Esteban Martirena y Pablo Bustos
Backup mentor: Ramon Cintas
3. RCIS: improving RoboComp simulator with transmission delays
RCIS is RoboComp’s robotics simulator. It is built using the OpenSceneGraph 3D engine as the visual renderer. Over the years we have tuned RCIS to suit or needs for robot software development. Despite the mature state of the software we need to add more functionalities to cope with the increasing complexity of our social robots. One of these new functions is the simulation of different kind of transmission delays that occur when the real robot is running. Ignoring these delay may even cause that a nice simulated result is almost useless in a real-world execution. The goal of this project is to design, model and include a set of realistic delay into RCIS so simulations get closer to real-world results. These delays will include transmission, hardware and computational factors.
Technologies involved: C++, Python, Qt5
Mentors: Pablo Bustos y Pilar Bachiller
4. Emotion recognition component for Learnbot
Learnbot is a small low-cost robot designed to develop computational thinking in kids of age 10 and above. Learnbot is programmed using a block language specifically designed for it. Through this language, kids program robot behaviors in an easy way by specificating what the robot has to to do whenever a given situation occurs. The last version of Learnbot has been equipped with a display that provides it with the possibility to show “emotions”. This way, kids can program not only motor behaviors, but also emotional reactions. To improve the HRI functions in Learnbot and extend the situations that make it to exhibit different emotional behaviours, we want to add new skills in the robot that allow it to recognize emotions in people such as happy, sad, angry, scared, … These new skills will be developed as a Robocomp component in charge of detecting people faces, recognizing basic emotions on them and providing its results to other components demanding this kind of information.
Technologies involved: C++, Python, Qt5, OpenCV3
Mentors: Pilar Bachiller, Iván Barbecho
5. Improving the Human-centered Robot Navigation Agent
Currently, RoboComp provides robots with the capability to behave in a socially acceptable manner. Particularly, the strategy followed in our Social Navigation Agent is based on the definition of regions where navigation is either discouraged or forbidden (personal spaces). These regions are related with the Proxemics, which defines spaces that humans mutually respects during an interaction. The navigation architecture combines the Probabilistic Road Map and the Rapidly-exploring Random Tree path planners and an adaptation of the elastic band algorithm to include the social behaviour. All of them are part of different RoboComp components. Considering the complexity of this task, the use of the current version of the social navigation agent is restricted, since it does not analyse, for instance, the human intentions during the robot navigation (do the human want to interact with the robot?) or the size of personal space depending of context (personal space should be different in a corridor than in a big room).
The task at hand here would be to integrate improvements in the existing RoboComp Social Navigation Agent. In a future version of this agent, more complex situations will be taking into account, such a s human intention analysis and/or adaptive personal spaces. A typical case of use would be that of a group of humans walking in different environments while the robot performs a task. All these simulations would be made using RCIS (RoboComp InnerModel Simulator)
Technologies involved: C++, Python, Qt5
Mentors: Pedro Nuñez, Luis Manso
6. Generation of new use cases, tutorials and reference information for RoboComp
During several years of RoboComp diffusion we have realized that it takes too much time for new users to understand the internals of the framework and its workflows. Creating tutorials of small use cases for RoboComp can help everyone to better understand all the tools involved and to facilitate the collaboration with our project in a much faster way. The student selected for this idea will have to acquire a fairly deep knowledge of the tool to select and create a set of key use cases that will help to explain and disseminate all the features available in RoboComp. These use cases must be documented and integrated in the RoboComp website and in the manual, so they are available to everyone and even to future GSoC students!
Technologies involved: C++, Python, Qt5, Jekyll (for static web generation) and other online generation tools.
Mentors: Marco A. Gutiérrez, Luis J. Manso
7. Gazebo-RoboComp connection
RoboComp currently has its own simulation tool RCIS (RoboComp Innermodel Simulator), however it does not incorporate physics simulation. Sometimes robot models are already available under Gazebo, which is a more common simulation environment. Having gazebo as an option for simulations with RoboComp will help grow the possibilities our framework can offer to developers. The idea is to connect RoboComp and Gazebo in a similar way ROS is. This way any simulation with RoboComp can be tested on both simulators (Gazebo or RCIS).
Where to start and what to include in my application?
- Download robocomp: https://github.com/robocomp/robocomp
- Get familiarized with the framework through the docs: https://github.com/robocomp/robocomp/tree/master/doc
- Run a test with the RCIS tool and understand how it works
- Download and run Gazebo: http://gazebosim.org/
- Run it and find ways to connect Gazebo to robocomp components through the ros interface.
Needed skills: Gazebo, C++, Robotics simulation
Possible mentors: Marco A. Gutiérrez, Ramón Cintas
8. Javascript support
An interesting diversion from current robotics technologies based on C++ and Python is the use of Javascript as the language to code some highly concurrent components. NodeJS component code generation was added during GSoC 2017. This was an innovative task as Zeroc ICE (the main supported middleware in RoboComp) just added support for javascript interfaces. Although components development was achieved some effort is still needed to stabilize the code and fully integrate this solution into the RoboComp component model. The RoboComp component generator will be extended to include Javascript running in a server as a new target language for components.
Needed skiils: Javascript, ZeroC ICE, Python, Cog(python module)
Mentors: Marco A. Gutiérrez, Nicolás González
9. Learnbot Simulation with RCIS
Learnbot is the new open robot supported by robocomp designed for learning. Although it is already in its second (or more) design iteration there is no support for this robot in the official robocomp simulation tool (RCIS). This is a key feature for the robot as it helps people test their algorithm in a simulated environment prior to real world testing while also giving the opportunity to developers to play with the robot without the need of the possessing the hardware.
In this idea the RCIS tool will be extended and different interfaces will be created so the LearnBot can be simulated. The behavior and shape should mimic the real hardware. Documentation and tutorials with different use case scenarios would be produced for future users and developers.
Needed skills: C++, 3D design, Zeroc ICE
Mentors: Marco A. Gutiérrez, Pilar Bachiller
10. Learnblock (Learnbot programming language) extension for collaborative robotics
Check: https://github.com/ibarbech/learnbot/blob/master/doc/learnblock.md
LearnBlock is the programing tool designed for easy usage of the Learnbot robot. It is meant to be used by students and to help them learn new concepts by the usage of robotics programming. This idea involves the extension of the Learnblock programming tool to manage collaborative robotics between learnbots. The student will have to develop a system that enables Learnbot to communicate between and offer this option through the learnblock tool to the end user. Finally a use case example of these collaborative robotics tool will have to be developed and tutorials produced for future users and developers of the platform.
Needed skills: C++, Python, Qt5 , OpenCV3, PCL Mentors: Marco A. Gutiérrez, Iván Barbecho
11. Adding locomotion skills to a human avatar
Real-time control of three-dimensional avatars is an important problem in the context of robotics. Currently, RoboComp provides a basic avatar which is used not only for simulations in RCIS (RoboComp Innermodel Simulator), but also for different navigation and human-robot interaction algorithms (fakehumanComp). RoboComp Human avatar implements basic locomotion skills, it moves from a keyboard and a joystick, but it doesn’t generate realistic walking motions. The task at hand here would be to integrate improvements in the existing RoboComp human avatar. In a future version of this avatar, more complex and realistic walking motions should be provide. Besides, social skills should be also taken into account in the final version of the component. All these simulations would be made using RCIS (RoboComp InnerModel Simulator)
Technologies involved: C++, Python
Mentors: Pedro Núñez, Luis Manso
12. Integrating MORSE Human-robot interaction simulator with RoboComp
MORSE is a generic simulator for academic robotics (see https://www.openrobots.org/morse/doc/stable/what_is_morse.html). Currently simulations in RoboComp are made using RCIS (RoboComp InnerModel Simulator), which has several limitations related to Human-Robot interactions. MORSE does not make any assumption on the external architecture using it, thus, the idea is to integrate this simulator as an additional tool for RoboComp. As MORSE authors say in their webpage, “it is developed using python and also supports a simple socket-based protocol for easy integration in other languages/toolbox”, features that will facilitate the proposed integration. Besides, MORSE developers provide complete bindings for Python which make it a natural way to integrate with RoboComp’s Python code generation tool.
Needed skills: C++, Python
Mentors: Pablo Bustos, Luis J. Manso
Backup Mentor: Pedro Núñez
13. Improvements in Robocomp code generator: robocompdsl
RoboComp’s code generator, robocompdsl, is one of the core elements of RoboComp. It is used to generate and automatically modify the software components that users develop. The code generated using robocompdsl is separated in two kinds of files: ‘specific’ and ‘generic’ files. The names of those files which are supposed to be written by the user end with ‘specific’, whereas those responsible aren’t end with ‘generic’. This separation allows modifying the specification of components and re-generate their code without overwriting user’s code. Over the years, it has been endowed with plenty of features: support for ROS, AGM, Qt4/Qt5, hierarchical state machines and much more. Despite it was initially developed using Java/Xtext, now it’s implemented using PyParsing/Cog (two Python modules). The tool is under constant evolution and we have multiple tasks to be done:
- Create special specific files for AGM and ROS, to help users structure their code.
- Update the documentation of the tool
- Update code generation to replace Qt with c++11 features
- Add support for Python 3
Needed skills: Python, Cog (python module)
Mentor: Pablo Bustos, Luis J. Manso
Backup mentor: Marco A. Gutiérrez
14. Graphical, automatic, real-time animation of RoboComp state-machines.
A very important tool of RoboComp is the state-machine code generator. The design and implementation of complex robot behaviour demands specific tools to create and debug highly structured programs that can be debugged during real-time execution. This project proposes the extension of RoboCom’s state-machine code generator to include new derived classes that will bring real-time animation of the program flow. The student will need skills in C++, Qt and Python to create code that will display the state-machine as an animated structure in a transparent way for the user of the tool. This graphic feature will be optional and non intrusive for the logic counterpart.
Needed skills: C++, Python, Qt5.
Mentors: Iván Barbecho, Pablo Bustos
Backup mentor: Ramón Cintas
15. A Natural Language Understanding (NLU) component for RoboComp
Significant efforts have been carried out in the Human Robot Interaction (MHRI) research field to provide multimodal and user-friendly interactive experiences between users and robots. As between humans, speech is the most natural, simplest and ideal modality to interact with robots using natural language processing technique, in particular when interacting with social robots.
Different Semantic Role Labelling (SRL) systems have developed for their use in different NLU-based applications. The main purpose of this project is the study of different SRL systems freely available, mainly those based in Deep Leaning tecniques. The libraries and APIs provided by these systems will be used and will have to be adapted for their use in the RoboComp cognitive robotic architecture.
As a result of the project, a component capable of understanding the natural language of speech input sentences will be developed. The student will also have to evaluate the quality of the semantic representation of knowledge, for the development of future dialogue systems between humans and robots.
Needed skills: C++, Python, Qt5. Mentor: Ismael García-Varea
16. An Automatic Speech Recognition (ASR) component for RoboComp
Significant efforts have been carried out in the Human Robot Interaction (MHRI) research field to provide multimodal and user-friendly interactive experiences between users and robots. As between humans, speech is the most natural, simplest and ideal modality to interact with robots using natural language processing technique, in particular when interacting with social robots.
Different ASR modules have been developed so far for their use in different speech-input based applications. The main purpose of this project is the study of different ASR systems freely available (offline and online). The libraries and APIs provided by these systems will be used and will have to be adapted for their use in the RoboComp cognitive robotic architecture.
As a result of the project, a component capable of recognizing the speech of different users will be developed. The student will also have to evaluate the quality of the captured signal and the transcriptions obtained, for the development of future dialogue systems between humans and robots.
Needed skills: C++, Python, Qt5. Mentor: Ismael García-Varea
17. Rewriting RoboComp’s QMat with Eigen
RoboComp’s core algebra library, QMat, was created before the Eigen library http://eigen.tuxfamily.org/ was so widely used and stable. Currently, QMat uses a QSharedData buffer based on the Qt library and provides a series of operators, mostly implemented in the QMat class itself. Eventually, Eigen became more popular and many widely used libraries such as OpenCV or PCL use Eigen matrices internally and in their API. Given that Eigen is now the de facto standard, it would be desirable to rethink QMat as a subclass of Eigen, so that we can work with these matrices used by so many libraries seamlessly.
The goal of this project would be to implement a new version of QMat as a subclass of an Eigen matrix. In the way, many methods will be removed from QMat because they will be inherited from Eigen. A lesser amount of them, related to geometry, will be kept. This new class should be properly integrated with another class of the framework named InnerModel, which would probably be affected by minor changes of QMat’s API.
Mentors: Luis Calderita, Ramón Cintas
18. Visual Detection Mechanisms in Mobile Robots
In this action the student will study and code a visual attention algorithm for RoboComp that will respond both to bottom-up cues and to top-down requests. Bottom-up cues are computed by a RoboComp adapted Yolo server that provides a list of detected objects and its probabilities. Top-down cues are provided by RoboComp’s internal representation of the scene (and the robot itself) and predict where objects are to be found after movement of the robot. The combination of both action streams will elicit an attentional behavior and a basic notion of self-awareness.
Skills required: C++ (advanced), OpenCV
Mentors: Pablo Bustos, Ramón Cintas
Mentors:
Marco A. Gutiérrez
marcogATunexDOTes
Robocomp Developer
Pablo Bustos
pbustosATunexDOTes
Professor, RoboLab,
University of Extremadura
Luis J. Manso
lmansoATunexDOTes
Postdoc Researcher, RoboLab,
University of Extremadura
Ramon Cintas
rcintasATunexDOTes
Researcher, Robolab,
University of Extremadura
Esteban Martinena
martinenaDOTestebanATgmailDOTcom
Robocomp Developer
Pedro Núñez
pnuntruATunexDOTes
Professor, RoboLab,
University of Extremadura
Pilar Bachiller
pilarbATunexDOTes
Professor, RoboLab,
University of Extremadura
Ivan Barbecho
ibarbechATalumnosDOTunexDOTes
Researcher, Robolab,
University of Extremadura
Nicolás González
nicolasATunexDOTes
Developer, Robolab,
University of Extremadura