 The U.S. Army launched this robotic collaborative technology alliance in the early 2000s just as robots began entering the battlefield. With the global war on terrorism came IEDs, and to combat them, the U.S. military teamed with industry and academia to tackle what would become its greatest challenge, replacing teleoperated robots with autonomous systems that could turn robotic tools into robotic teammates. Robots depend on our robots being precise, being accurate, and being dexterous. Through focused research pursuits, the Army is advancing research to solve some of the most difficult problems facing allied forces. By collaborating with America's top academic institutions, the Combat Capabilities Development Command Army Research Laboratory is pushing the frontiers of scientific research to help robots autonomously perceive, plan, and interact. The military and the Army context means that you can't know the environment ahead of time. It's going to change. Much of today's advancements in robotics, autonomy, and artificial intelligence is credited to the foundational research pursued by America's military and its university and industry partners. Strategic investments in defense-led foundational research has resulted in advanced science and four critical areas of ground combat robotics that affect the way U.S. warfighters see, think, move, and team. Academia is all about solving the most challenging problems out there. This collaboration is about developing the algorithms that look at the problems in a basic shape and form, find the algorithms that solve those problems, and then transitioning those algorithms into the technologies that are much more mature. And this is where the Army and collaboration with industry really helps, because researchers on the Army side can actually take those algorithms and actually build the technologies that can solve the problems using the algorithms that were developed on Academia. Here at MIT, we've tried really hard to provide value to the Army Research Lab through the various programs and by working with the researchers. We developed new forms of planning algorithms. We developed new forms of algorithms that allow robots to team naturally with human partners. So for instance, I and my research lab with my students and my postdocs, we've developed a number of systems that allow a robotic wheeled vehicle or a robotic mobile manipulator to actually take instructions from a human operator or a human partner, carry out complex temporal extended tasks over long distances in order to meet the needs of the warfighter. And so we're developing new kinds of algorithms, we're developing new models, and we're showing how these systems can perform well, can perform at army tempo, at op tempo, without the training data that most existing systems assume. I think I have been involved in the RCT program for six years. I think this is a very rewarding appearance for me because it allows me to interact with people, especially talented scientists and engineers from the different places. So my work focuses on a certain component which is critical in autonomous systems, but the interaction and collaboration with other experts allows me to have a more global understanding of the entire autonomous systems and I also learn from them. Being in a collaborative team I think really helps because one person cannot do all these things, right? And you also get to learn from people who are actually experts in these different areas. And I think that's one of the major reasons why I think collaboration is very useful. I have neither faculty, I am neither a student, so I am a person that works in between the faculty and the students. I have had the opportunity to see these students on the academic part in the classroom with faculty and then developing their approaches in the lab and then taking them to the field to work on real problems on the field with actual robots and that is an invaluable experience. These very same students also have the opportunity to interact with, for example, soldiers in the field or with colleagues from the ARL that have very different opinions and insights into what is relevant in the field of robotics. It was a great opportunity to test robots and environments that you don't usually get to use. As a university we're constrained to drive around in our hallways or around on campus and that's a very limited set of environments that don't reflect the challenges of the real world and Campbell June gave us an opportunity to go out and try more difficult things and see where the state of the art fails and where it succeeds. The RCTA project has been invaluable in helping us take some of the research that we're doing and partner with researchers and engineers in the Army and put it out in the real world to see how it performs. Some of the lessons learned are very harsh. The algorithms that I thought I had created that would work really well out in the real world ended up not performing well at all when subjected to the harshness of the real world. So I think what's really valuable is in having problems that are real, having problems that are meaningful, having clarity in metrics and partnering with people who have the same vision that we have of getting robots into the real world solving real problems. The collaboration with RCTA or in general with other researchers really highlights different perspectives of looking into the same problems and also by working with RCTA I get to see different types of problems that I can't really think of on my own when I'm researching in this perfectly safe environment. The students benefit from exposure to real world problems. They benefit from exposure to problems they wouldn't necessarily see in other domains and other disciplines. They benefit from the rigor that is brought to these programs. They benefit from being able to collaborate with other researchers and they also benefit technically from being able to advance the state of the art. The undergraduate and graduate students in my lab and in my collaborators labs, the opportunity to work on the highest priority problems of national defense has represented opportunity to get exposure to problems they wouldn't necessarily see elsewhere. Knowing how to get rid of a lot of the assumptions that exist in a lot of commercial systems and put them into systems for the military context is hugely valuable for the students. I got to visit the Army Research Lab and hear a lot more about what was going on and have like ongoing dialogue with the researchers where they would pose problems and questions that I would then go off and they would shape my research direction. So I was also funded under the RCTA during my faculty career. We're able to educate an entire new generation of PhD students, master students, undergrads, young faculty to think about what it is to solve problems out in the real world. So I think that we're sort of sowing the seeds of trees that will grow and really be the future of robotics for the United States. The Army Research Lab has contributed substantially to the next generation of engineers in industry and academia. A lot of industrial results are really dependent on the PhDs that are growing inside the collaboration with the Army Research Lab. I believe for students it's really beneficial to not just work on their small isolated projects but really see the bigger picture, working on interdisciplinary teams and I think that's also of course going to be really crucial later on when they go into the workforce and that is also of course exactly the kind of lessons they learn through participation in programs like the RCTA. Many of my students have gone on to academic careers, many of them have gone to found their own companies, many of them have gone to work in large defense contractors as well. This is all tremendous benefit to them and hopefully benefit to the nation as well and they're working on the next generation of problems. I just actually graduated this year and I will be joining Google DeepMind. This is a group that actually does like artificial intelligence research so I'll be doing basically AI research. I'm going to essentially continue my PhD work forward so still again working with manipulation, working on like perception and trying to sort of get robots into unstructured environments. I came to the University of Pennsylvania as a master's student but during my time on the RCTA I was accepted into the University of Pennsylvania as a PhD program in large part due to my work on the program. I feel very fortunate to have had the opportunity to work with the great people Army Research Lab, people like Dr. Stuart Young, people like Dr. Ethan Stump, Dr. John Fink and many others. A lot of the people at the Army Research Lab represent the best and the brightest of what our nation has to offer. They bring tremendous intellectual heft and rigor to the problems that we're tackling and they represent a real commitment to keeping the nation safe in a variety of ways. I think the Army Research Lab has gone unrecognized a lot of the time for its contributions in developing basic technology and transitioning that technology out into the world. Through this program there have been a lot of folks who graduated have gone to all the top companies that are building robotic systems. From my lab I have placed students at Uber, Apple, Amazon Robotics and a number of those companies and they're bringing in those kind of knowledge and expertise from the programs like the RCTA into the industrial world. The RCTA has given me better appreciation for the challenges of field robotics, moving robotics out of like a research lab and into the real world. My father was in the Army so I definitely knew there were talented people in the Army, talented engineers in the Army but what I didn't know was that there were talented researchers. I suppose I was sort of vaguely aware that the Army must be researching the future of war but I thought most of the hands-on engineering research would be going on in defense contractors' labs and so this collaboration with the Army Research Lab really highlighted that no that's not the case and that a lot of the most interesting research is being directly supported by the Army and by civilians working with the Army. I think the RCTA is a tremendously exciting project and it was a privilege to be a part of it. I learned a tremendous amount and I hope the Army funds similar things in the future and that I can be part of it.