Speaker Interview: Dr Blaine Levedahl, Director, Government Relations at Olis Robotics
We’re delighted to introduce you to Blaine Levedahl, director of government relations at Olis Robotics. Blaine will speak at Space Tech Conference on Wednesday, May 22, discussing artificial intelligence and autonomous systems.
Blaine holds a BS and an MS in aerospace engineering, a MS in electrical engineering, and a PhD in aerospace engineering. He has 16 years of experience developing modeling strategies and control systems for a myriad of different projects, from collaborative control of autonomous unmanned aerial vehicles, to rapid maneuvering control for underwater vehicles, to path-planning and control of single and multiple robotic systems. At Olis Robotics, Dr Levedahl serves as director of US government programs, where he directs the application of the company’s core telerobotic technologies to address customer needs, including use in planetary exploration and in-space servicing.
Hi Blaine, could you please tell us a little bit more about your position as director for US government programs at Olis Robotics?
In my role as director of US government programs, I orchestrate the productization and application of our core remote robotic manipulation software in space, underwater, and terrestrial domains. This includes working with customers to identify their needs, architect solutions that allow our core technologies to meet those needs, as well as define strategies and roadmaps that help influence the direction of our core technology to ensure we remain competitive.
You will participate in the ‘Optimizing Mission Agility with Autonomous Systems: The Benefits and Challenges of Implementing Artificial Intelligence’ panel session on the Conference on Wednesday, May 22. The terminology of autonomous systems, robotics, and artificial intelligence sometimes becomes jumbled. What would be the best way to describe autonomous systems for the space industry?
In the space industry robotics gives humans the opportunity to explore remote places human beings can’t currently reach. It also provides an excellent way to overcome the inhospitable nature of space in places humans can venture. For example, Extravehicular Activity (EVA) operations aboard the International Space Station (ISS) have seen an increase in the number of EVA tasks being performed with remote robotic manipulation.
These robotic systems generally have different levels of autonomy. The level of autonomy used in the space industry is usually defined based on the operational environment. For example, on distant planets, where the communication time delays can be ten minutes or more, systems will typically use as high a level of autonomy as practical within the uncertainty of the environment and complexity of the task. This can be contrasted with robotic operations on the ISS where more human interaction is possible.
Artificial intelligence can potentially provide a means to increase the level of autonomy of robotic systems. In the space industry, however, the missions are typically costly and possess high levels risk, particularly if they involve human life. This has led to a justifiable need to perform comprehensive verification and validation testing prior to system deployment. Performing verification and validation testing of systems with artificial intelligence, in which autonomous actions are unbounded or bounded but depend on an extremely large number of parameters that need to be exercised in testing, presents a dilemma that has not yet been reconciled.
What are the challenges in terms of implementing autonomous systems on either the manufacturing floor or within space systems?
In my opinion, the challenges of implementing autonomous systems are related to the complexity of the operational environment and the operational tasks being performed. As the level of complexity of the environment increases (for example, a more cluttered or dynamic environment) or the complexity of the tasks increases (for example, performing operations with tools designed for human hands), the complexity and sophistication of the perception and decision making of the system to perform the tasks autonomously increases. As the perception and decision making of the autonomous system becomes more complex, so does the verification and validation testing to ensure correct, reliable, and safe operation. The challenge, I believe, lies in developing sophisticated perception and decision-making capabilities and performing verification testing for autonomous system that can perform complex tasks in complex environments. These are the challenges that OlisOS is addressing from the software side with Progressive Autonomy.
Attempts to develop autonomous systems for increasingly complex environments and tasks is ongoing in both the terrestrial and space domains. In the terrestrial domain, we have already begun to see the transition of autonomous robotic systems to perform more complex tasks in more complex environments. For example, autonomous robotic manipulators that previously supported simple repetitive tasks in structured environments – such as auto body welding in an assembly line on manufacturing floors – are now being used in more complex environments and with more complex tasks such as the pick-and-package from a pool of candidate items task. In the space domain, there is generally less control over the environment, which leads to unique challenges. The drastic sun and shade scenarios cause complications for the perception system. The fact that the objects and the manipulation platforms are ‘floating’ in zero-G requires decision-making capabilities on how to approach and capture an object. And the limited robotic hardware that can meet the temperature and radiation requirements restrict leveraging autonomous systems implemented in the terrestrial domain.
What will autonomous systems mean to the space industry in the coming decade?
Autonomous robotic systems are an enabling technology for the space industry. Autonomous robotic systems will enable in-space servicing, assembly, and manufacturing, which will drive down the cost of maintaining payloads on orbit as well as enable further human exploration of space. For this to happen, I believe that the space industry will need to become increasingly comfortable with and begin to leverage autonomous systems more in the coming decade. I believe we will first see increasing levels of autonomy demonstrated to service on-orbit assets in low-Earth orbit (LEO) both internal and external to habitats. This will likely be the first step because it is less costly, correct operation is easier to verify, and we already have existing infrastructure such as the ISS and multiple low-cost launch providers that can be leveraged. I believe autonomous robotic systems to support human exploration of space will not be too long after and will take advantage of the autonomy demonstrations in LEO. As we set our sights past LEO to more inhospitable destinations like the proposed Deep Space Gateway (DSG) and the moon, we will need to utilize autonomous robotic systems to establish landing zones, prepare for human arrival, and perform caretaking tasks when humans are not present.
In terms of industry news, what development, announcement, or otherwise has stood out most to you in the past year and why?
Over the past year, I have been very excited by the increasing interest in remote robotic capabilities for space applications, such as on commercial space stations like the NanoRacks Outpost Program; on human exploration missions such as the Deep Space Gateway; and for in-space servicing such as on NASA’s Restore-L mission. I am also very encouraged to see organizations like CONFERS being established to help support remote robotic operations in space and define operational standards.
We’re looking forward to seeing you at the Conference at Space Tech Expo. Can you tell us what you’re most looking forward to at the show?
I believe the current infrastructure in LEO and the decreasing launch costs provide excellent opportunities to demonstrate robotic capabilities in space. I am looking forward to talking with others at the conference and exchanging ideas about how to take advantage of this infrastructure to forward robotic operations in space.
Want to learn more about autonomous systems and artificial intelligence for our future spacecraft? Make sure to join us at Space Tech Conference on Wednesday May 22, to hear Blaine speak alongside other speakers from University of Southern California, Raytheon, The Aerospace Corporation and NASA Ames Research Center. Make sure to book your pass here!