Picture this: You purchase a new robot and spend countless hours figuring out how to get it to move. Finally you have it working enough to begin developing software and features to further your research. As you work, you have new ideas, the research expands, and you think of additional capabilities and programs you’d like to try. Unfortunately, you’re stuck if the core operating system isn’t open to modifications because expanding beyond the original robotic system capabilities is often not possible. You either have to go back to the manufacturer and hope they grant you access to their software, or start over with a new robotic platform or system.
Reinventing the proverbial wheel each time a researcher wants to try something new wastes time and effort, slowing down innovation in the field. Fortunately, a solution exists. Open-sourced systems changed the game for researchers by tackling a major obstacle in robotics development: The need to code software for robotic functionality and integration from scratch.
One system in particular is gaining traction over proprietary robotics software. ROS (Robot Operating System) is an open source, BSD licensed, flexible system for writing robot software that offers users a way to quickly build, maintain and expand their robots’ capabilities. This collection of tools, libraries and conventions provides standard operating system services that enable control of low-level hardware. Unlike most proprietary software—which is restricted in how it can be customized or integrated with other technologies—ROS allows researchers to develop add-ons and supplemental controllers for a wide range of tasks.
In the last few years, ROS robot availability expanded. Now the ROS community includes many different hardware systems from telepresence robots to skilled mobile manipulators. By connecting hardware to software in a simplified way, ROS provides the architecture for navigating, sensing and manipulating. Quite simply, ROS enables researchers to build more complex robots in a much shorter amount of time. For example, ROS robots have enabled quadriplegics to explore their world again, performed household chores like the dishes, helped astronauts perform experiments in space and much more.
Traditionally, robotics development is not an easy field to work in. For example, a researcher developing a home assistant to pick up specific objects in a cluttered room would need to develop hardware and software from the ground up. With a robot controlled by proprietary software, the researcher must code the robot’s basic actions from scratch, and also risks not having the adaptability and flexibility needed to expand the robot beyond its original intended use.
ROS is the first widely-used open source robotic operating system to support research platforms. ROS offers a shortcut for researchers to quickly equip robots with basic software capabilities so they can focus on more meaningful experimental work. Universities, startups, government and, recently, bigger companies have developed robust functionality and integration across a spectrum of robot platforms by leveraging work published by the ROS community.
ROS first emerged as framework prototypes at Stanford University before the robotics company Willow Garage formally developed ROS in 2007. ROS provided standard operating system services (e.g., hardware abstraction and package management) and offered basic controllers for mapping, localization, navigation, manipulation and more to a growing number of researchers. In February 2013, an independent nonprofit organization called Open Source Robotics Foundation (OSRF) took stewardship of the core development and maintenance of ROS, with the mission to “support the development, distribution and adoption of open source software for use in robotics research, education, and product development.”
In short, ROS, with its minimal constraints for users, is going strong, with an ever-growing and diverse community. By sharing code freely, collaborating and supporting each other, researchers collectively provide tools that enable ROS robots to do more and more.
The ROS community’s collaborative nature is vital, as members specialize in different areas and publish work for others to leverage. For example, ROS programs exist to recognize object edges or to carefully grasp fragile objects. Currently ROS is on its 8th official release, dubbed Indigo Igloo (supported on Ubuntu Saucy and Ubuntu Trusty). As an increasing number of developers, students and research groups share their work, new research builds on old. Now, the ROS community is expanding beyond universities to commercial and industry applications.
While a flexible, adaptable and open source system sounds great, one might wonder, what are the disadvantages of ROS robots?
As with any framework, there is a learning curve. Researchers new to ROS may find the scope overwhelming at first, but the advantages outweigh the cons as evidenced by the diverse and rapidly growing ROS community.
For example, lack of formal troubleshooting support from a manufacturer is made up for by an active and collaborative community with how-to wikis abound. Once developers get their bearings and begin to take advantage of the flexibility of ROS, they become freed up to do what they really want to do: use robotics to solve complex issues and develop solutions to real-world problems. ROS is helping robotics make significant advances in healthcare, home assistance, education, transportation, manufacturing and many other industries.
Increasingly, a number of companies are starting to offer ROS robots—Stanley Innovation included. We offer a host of customer-centric robots that can run on ROS and are customizable to a wide range of applications. Rather than buying a robotic platform and spending costly time figuring out simple navigation and integration before even beginning the really crucial work, users can now buy robot platforms that utilize the rich ROS resources and provide shortcuts to faster innovations.
The Segway Robotic Mobility Platforms (RMPs), an immensely popular and robust system that serves a number of industries, did not have a core operating system open to development. However, since our founders are former Segway innovators, Stanley Innovation is the only company with the license to access the Segway core operating system, and our engineers have developed a line of RMPs with various levels of ROS capabilities out of the box. Our engineers can very quickly integrate components to your existing RMP—or even develop a custom RMP—to work with ROS and leverage software published in the ROS community. With a redesigned system architecture, robot platforms are customizable to a range of applications. Users can purchase an out-of-the-box solution meeting their requirements and focus their efforts on more compelling research challenges. For example, a recently released version of RMP, called the RMP220RE, is a ROS-enabled robot with potential for a wide variety of research and industry uses.
As more robots become open, flexible and compatible, roboticists will be able to more rapidly solve interesting problems and answer big questions, building upon previous work to achieve meaningful results. And as more and more companies begin to offer ROS robots and integration, paving the way for significant breakthroughs, we may finally see the field of robotics reach its potential to help advance humanity and, truly, change the world.
If you’re interested in talking with our engineers about how a ROS robot can advance your next robotics research project, contact us today. We would love to transform your research ideas into reality.
Get News & Resources Delivered to your inbox.Sign Up