ROS Bootcamp

Robot Operating System (ROS) Bootcamp Funded Training Opportunity

We invite experienced C++ programmers to come for an intensive 3-day introduction to the Robot Operating System or ROS.  Funding is available for some attendees.

Download the Application Here

Activities and Format

ROS Bootcamp will consist of three full days of intensive instruction broken into six half day sessions.  Each half day session will begin with a brief lecture and demonstration and then be followed up by several hours of hands-on, closely supervised learning activity where students will apply what they have learned to solve realistic problems using an eduMIP robot (https://www.renaissancerobotics.com/eduMIP.html).

Students are expected to attend and fully participate in all six sessions as they are cumulative.

When and Where?

ROS Boot camp will be Nov 29, 2017 – Dec 1, 2017 in Clark 507.  Exact times TBA but plan on all day.

Why?

In addition to comprising a ‘Robot Operating System’ ROS software packages have the potential for being very useful in a wide spectrum of oceanographic instrument development efforts. In addition to immediate project needs several groups at WHOI will be exploring the idea of more wide-spread use of ROS as the way for oceanographic instrument developers (scientists and technical staff) to support each other’s R&D efforts and share design, code, and ideas more fully across departments and disciplines at an institutional or even multi-institutional scale.

WHO

This course is open to all active WHOI employees and students who meet the pre-requisites. Pre-Registration via the application process is required even if you are not requesting funding.  Funding including materials and salary support is available for approximately twelve attendees.

Pre-Requisites and Pre-Course Homework

The pre-requisites below were established after consultation with the instructor.  Questions, clarifications, or requests for exception should be discussed with Carl Kaiser (ckaiser@whoi.edu x3269) who will consult with the instructor if necessary.

  • An intermediate understanding of C++ (generally 2 courses, or a few months of intensive experience should suffice)
  • Regular use of modern version control (GIT and Mercurial)
  • Command-line knowledge of Linux.

 

In order to make efficient use of the instructor’s and fellow students’ time, all participants will be asked to complete several tasks ahead of time including:

  • Build up a laptop with Ubuntu 16.04 LTS and ROS Kinetic – No virtual machines, native or dual boot only
  • Assemble an eduMIP robot kit (https://www.renaissancerobotics.com/eduMIP.html) – kit and 32GB Micro-SD card provided for those who are funded – those not funded will need to provide their own although a limited number are available for JP students or post-docs on a first come first serve basis.
  • Install a pre-built Ubuntu 16.04 LTS linux image on a 8GB (or larger) Micro-SD card, including beaglebone blue and robotics cape software and ROS kinetic, as described here https://dscl.lcsr.jhu.edu/home/courses/EduMIP_ROS and work through the sections.
    • Assembling and Testing the EduMIP Kit
    • Setting Up Your Beaglebone Blue Wireless Board for Your EduMIP
  • Complete several tutorials TBA at http://wiki.ros.org/ROS/Tutorials

 

Participants who are funded are expected to receive approximately 4 hours of funded preparation time to complete these tasks in addition to their time for the course.  It is likely that this may not entirely cover the total prep time required.

Registration/Funding

Pre-registration is required for all participants regardless of funding source.  Please register by completing the form on the last page of this announcement.  All registration forms should be e-mailed to ckaiser@whoi.edu no later than 5pm EST October 30, 2017.  Confirmation of attendance and award of funding is expected no later than Monday November 6, 2017.

Attendance Criteria

All attendees meeting the pre-requisites are expected to be accepted into the class.  In the unlikely and extreme circumstance that so many people are interested that the quality of instruction would be affected, it may be necessary to limit attendance, in which case the same criteria as used to select funded attendees will be used to limit attendance, but this is considered unlikely.

Funding

It is anticipated that funding is available for 12 people to attend.  The exact number of funded attendees will vary slightly based on funds available and the cost of the people selected.  Selection will be performed by a small committee of senior technical staff and will be based on the application form on the last page of this call.  The selection criteria are:

  • Does the applicant meet the pre-requisites?
  • The funds for this activity derive from the “Technical Staff Training and Development Opportunities” opportunity and thus technical staff will generally have priority
  • Attention will be given to diversity of department, lab, application, and seniority with the goal of getting broad participation
  • The immediacy of any use or potential use of ROS
  • The degree to which the individual is routinely involved in software development or decision making about software development for their work at WHOI
  • In close calls the degree to which a person is considered to be a lead adopter or trend setter may be considered

Background

Robot Operating System (ROS) is a rapidly expanding ecosystem of tools related to robotics, automation, control, and data processing.  Fundamentally, ROS is a middleware; a communications layer formalizing the interconnectivity of various portions of a robot control system.  However, ROS has become much more than just a middleware.  ROS has become a philosophy and an open source toolbox that is rapidly becoming the de-facto standard in research and academic robotics.  Still more importantly, ROS is becoming a community and can provide a common language that scientists and technologists can use to talk about a wide range of instruments under development.  There are many developers all working on tools which are essentially compatible and the official portion of ROS, that is the middleware keeps everything glued together making integration into a new project faster and easier.  Undergraduates using the ROS ecosystem are now completing Autonomy projects that less than ten years ago would have been major research projects at top tier universities.  ROS is by far the largest such community.

Multiple groups within WHOI are now beginning to embrace or consider use of ROS.  Select projects from OSL, DSL, and AEL in AOP&E are starting to use ROS, at least one Biology project is looking at it strongly, and the opportunities both within robotics and in related fields such as sensors, samplers, and real time data processing can all benefit.  In addition to availability of external tools, hw/sw tools developed within WHOI, aimed at common areas of ocean instrumentation need, could be more easily shared between groups if based on ROS.  Moreover, since most schools are now teaching ROS in robotics programs new college hires will be able to transition to WHOI more quickly, and offer new knowledge about ROS capabilities and use for use in the ocean environment.

Instructor

Louis L. Whitcomb, a Professor of Mechanical Engineering and Computer Science at The Johns Hopkins University, and WHOI Adjunct Scientist and collaborator.  He has been heavily involved in Jason, Jason II, Nereus, NUI, and numerous other WHOI vehicles.