GSOC 2017: Proposing new Library: Scalable Interactive Occupancy Grid Library

anindex

Hi everyone, I am An Le Thai, a second-year undergrad student of Vietnamese-German University. I would like to propose a Library that support researchers and hobbyists in creating 2D Occupancy Grid map. User can change the environment easily just by clicking on the map. There are many publications on active research fields such as Cognitive Perception and Motion Planning nowadays. Researchers have hard time to create a simulation engine to validate their works. The idea of this project is to create an unique, interactive and scalable occupancy grid map with various options to display, functionalities and data logging engine. This project will save a lot of time of developing, debugging and retrieving test program for researchers and let them focus on developing mathematical models and algorithms.
To test the environment, i also implement Maze Generator using Recursive Backtracking, search-based planning algorithm such as A*, D*Lite and DLR. These algorithms are the foundation of Motion Planning in Robotics. Any other custom generator can be built and integrated easily in this environment.

Here is the demos:

With this project, Processing Foundation can extend its reach to a large amount of researchers, who are working actively in Motion Planning and Cognitive Perception field. It also provides Processing Software ability to have an user-defined and interactive scalable environment that Processing currently do not have. The library will be the first scientific library that Processing will be able to target new parts of the community.

If you have any concerns, please let me know. These suggestions would valuable for me to create better proposal. Thank for your time.

Cheer!

anindex

I am looking for feedback on my idea. Any advice, comments, etc. are welcome.

Lord_of_the_Galaxy

(You won't like this, so maybe you shouldn't read)

Users of such complex software are unlikely to really need Processing. I suppose there must be some good reason why the most common users of Processing are creative/visual artists and not really "scientists"(in the more traditional sense). Nonetheless, the idea may be nice for those who actually want it. Maybe if you gave some sort of sample code/software, we could test it and rate it based on user friendliness, etc.

anindex

Many thanks for your comment.

To an extent, I agree with you. From the beginning, the idea of Processing is, providing an friendly environment for "teach programming fundamentals within a visual context". However, "Processing has also evolved into a development tool for professionals" and "The Processing community has written more than a hundred libraries to facilitate computer vision, data visualization, music composition, networking, 3D file exporting, and programming electronics". These quotes are on https://processing.org/overview/

There are many other examples that researchers and students take advantages of friendliness and alternative for restrictive simulation software for fast prototyping their projects . I refer this on the Education and Research part of https://processing.org/overview/.

Motion Planning is a very important field in Robotics. My idea concept is an universal grid map with conceptual robots to create a common environment that researchers can test their algorithms directly and even students can practice their learning. (Maybe i am wrong but i do not see there are any library with Motion Planning concepts yet) There is one important in the library that it will have user-defined data logging engine. Researchers have hard time to design a test program for their algorithm. I believe this library will reach many target in academic part of the community.

Here is the source code that you can test, it is just a prototype now. https://github.com/anindex/Processing-path_planning

For more information about my idea, this is my proposal: https://docs.google.com/document/d/1piIO9qe0IgaQcBVi8EiRh42pzqeUo-JmdifkKGNH2JI/edit?usp=sharing

Again thank you for your feedback. All suggestions are welcomed.

Lord_of_the_Galaxy

And again, I forget that robotics is no longer confined to large research institutions, but rather has started to become a common hobby. Maybe in the next 5 years or so, this project will have a much larger audience.