Overview of Research at UTRC's Systems Department
Dr. Claudio Pinello - Associate Director
Dr. Zamira Daw - Staff Research Scientist
Dr. Julian Ryde - Staff Research Scientist
Dr. Amit Bhatia - Staff Research Scientist
Host: Rance Cleaveland
This presentation will give a broad overview of research at UTRC’s Systems Department, with particular focus on the areas of robotics, autonomous systems, model-based design and formal methods. Autonomous and intelligent systems research for aerial and ground robotics includes intelligent system architecture, human-machine systems, perception, and collaborative motion planning with dynamic collision avoidance, manipulation, and formal verification. Research on model-based design and formal methods in UTRC focuses on the adaptation and application of theories and technologies in an industrial context to improve the development of our products. The presentation will conclude with a discussion of existing and future career and internship opportunities in the area of robotics, model-based design and formal methods.
The Cyber-Physical Systems group, based in Berkeley CA, will be highlighted, including a more detailed discussion of the RenderMap technology "Exploiting the Link between Perception and Rendering for Dense Mapping" and “Formal modeling of certification processes.”
RenderMap technology "Exploiting the Link between Perception and Rendering for Dense Mapping"
We introduce an approach for the real-time (2Hz) creation of a dense map and alignment of a moving robotic agent within that map by rendering using a Graphics Processing Unit (GPU). This is done by recasting the scan alignment part of the dense mapping process as a rendering task. Alignment errors are computed from rendering the scene, comparing with range data from the sensors, and minimized by an optimizer. The proposed approach takes advantage of the advances in rendering techniques for computer graphics and GPU hardware to accelerate the algorithm. Moreover, it allows one to exploit information not used in classic dense mapping algorithms such as Iterative Closest Point (ICP) by rendering interfaces between the free space, occupied space and the unknown. The proposed approach leverages directly the rendering capabilities of the GPU, in contrast to other GPU-based approaches that deploy the GPU as a general purpose parallel computation platform. We argue that the proposed concept is a general consequence of treating perception problems as inverse problems of rendering. Many perception problems can be recast into a form where much of the computation is replaced by render operations. This is not only efficient since rendering is fast, but also simpler to implement and will naturally benefit from future advancements in GPU speed and rendering techniques. Furthermore, this general concept can go beyond addressing perception problems and can be used for other problem domains such as path planning.
Formal modeling of certification processes
The approach of formal modeling of certification processes uses automatic reasoning in order to optimize the development process and to support business decision making while ensuring compliance with certification standards. The approach consists of combining formal models of the certification standard (what has to be satisfied), and models of the actual development process (how it is satisfied). A contract-based language is used to model the certification process. A SMT solver (Satisfiability Modulo Theories) is used to verify whether the certification is satisfied and to find optimization possibilities. The proposed approach is demonstrated using a case study to model a subset of the certification of a real aircraft cooling system certified under DO-178C. It is anticipated that modeling of the certification standards demonstrated in this case study would help meet the current challenge of creating new standards to certify new technologies.
Dr. Claudio Pinello is the Group Leader for Cyber-Physical Systems, based in the UTRC Berkeley California office. He is responsible for the development of talent and capabilities in the areas of Model Based Design and Autonomous and Intelligent Systems. Since joining UTRC in 2009, he has performed research and engineering work in the areas of simulation for distributed embedded systems, application of formal methods to aerospace systems, reuse strategies and design space exploration.
He holds a PhD in EECS from UC Berkeley. Previous research areas include automotive engine control, fault tolerant distributed systems, schedulability of distributed real time systems. He is co-inventor on one patent, received two best paper awards, the 2007 SAE Arch T. Colwell Merit Award, and a UTC award for furthering the dissemination of Model Based Design Technologies to one of our Business Units.
Dr. Julian Ryde is a Staff Scientist, Intelligent Robotic Systems, at the United Technologies Research Center in Berkeley, CA. His research focuses on perception for mobile autonomous systems particularly, localization and mapping from ranging and imaging sensors. In addition he is also investigating perceptual simulation for enhanced autonomous intelligence. Prior, he was a Research Assistant Professor in the Computer Science and Engineering Department at SUNY Buffalo where he lectured in Computational Vision and worked on perception for mobile autonomous systems. After his PhD he joined the robotics group at CSIRO’s Autonomous Systems Laboratory in Brisbane (Australia) in 2008 as a Post-Doctoral Research Fellow. There he was involved in 3D perception and mapping for mining equipment such as electric rope shovels and autonomous skid steer loaders. One of his principal accomplishments was autonomous soil moving and manipulation with a skid steer loader (bobcat). His PhD was in Cooperative 3D Mapping for Multiple Mobile Robots and was from the University of Essex (UK). He has BA, MSci and MA degrees in Physics from the University of Cambridge (UK). He has authored over 35 peer-reviewed journal, conference publications and book chapters. He was awarded best conference paper at the International Conference on Robotics and Biomimetics in 2006. His research interests include mobile robot sensors, 3D mapping, multi-robot collaboration, video processing and autonomous system perception.
Dr. Zamira Daw is a Staff Scientist in Model-based design, at the United Technologies Research Center in Berkeley, CA, since 2015. She has worked in projects related to model-based design for system development, for robotic multi agent systems, and for workflow certification processes. Previously, she was a postdoctoral fellow during 2 years at University of Maryland in the Institute for Systems Research and in the Computer Science Department. Her research focuses on the integration of formal verification into model-driven development processes; most of her work is at formal semantics, formal verification, model transformation, and code generation from UML models. She received her PhD in Electrical Engineering and Computer Science from University of Kassel, Germany in 2013 and an MS in Information Technology from Hochschule Mannheim – University of Applied Science, Germany in 2008. She also earned a BA in Electrical Engineering from Pontificia Universidad Javeriana, Colombia in 2004.
Dr. Amit Bhatia is a Staff Scientist with the Cyber Physical Systems Group at UTRC in Berkeley, CA. His areas of expertise are decision making under uncertainty, robotic planning and perception, formal methods, control theory and safety engineering. In his current role, he is responsible for development and deployment of novel AI and robotics-based innovations in multiple UTC product portfolios. Prior to joining UTC, he worked in oil and gas industry at Schlumberger Technology Corporation developing products related to equipment control, automation, prognostics and health management. As part of his masters, doctoral and postdoctoral research at UIUC, UCLA, MIT and Rice, he developed novel algorithms for solving task and motion planning problems, and problems related to safety analysis and efficient bug discovery for embedded control systems. His research has been cited over 400 times in various peer-reviewed conferences and journals.