Processor Design Exploration for Vision Based Mobile Robots

The following Great Innovative Idea is from Christopher B. Harris, the Presidential Diversity Postdoctoral Fellow in the School of Engineering at Brown University. Harris presented his poster, Processor Design Exploration for Vision Based Mobile Robots, at the CCC Symposium on Computing Research, May 9-10, 2016.

The Idea

At Brown University, we’re designing robots! Actually, we’re building the simulation tools used to design mobile robots. Mobile robots require specialized algorithms for tasks such as perception and control. These algorithms are often very processor intensive. As a result, much of the innovation in mobile robotics currently occurs at the software level. Robotics researchers will often devise new algorithms and run them on whatever computing platform is available from their local retailer.

At Brown, we are building design tools which will allow the creation of the next generation of embedded computing platforms for sensor-based mobile robots. Our platform combines models of computer processors with physics based robot models, actuator models, and sensor models to allow the simulation of complex, heterogeneous robotic systems. You can mix and match different sensors such as cameras, laser rangefinders, and inertial measurement units (IMUs) with various mobile robot designs and different embedded processors. You can even model new processor architectures that are specially designed to accelerate commonly used algorithms such as navigation or feature detection and see how they perform with a specific set of sensors and actuators while operating in a 3D virtual environment of your own design.

Impact

Computer architects have learned that computers are SYSTEMS and that both the hardware and software (and how they interact with one another) are important. As computers become more complex and begin to incorporate more diverse components, it is important to be able to accurately model these systems. Processor architectures that are well matched to the software algorithms, sensing capabilities, and actuators of a mobile robot will result in robots with greater autonomy and the ability to complete more complex tasks. Our design tools will allow engineers and scientists to optimize across subsystem boundaries to design more compact, higher performing, and more power efficient mobile robots. We hope that our platform will help to enable a new wave of innovation in hardware design for cyber-physical systems in general and mobile robots in particular.

Other Research

I am in the business of building better computers. What that means depends on the specific application as driven by the changing needs of society. My focus tends to be on the hardware and so my research areas include computer architecture, digital system design, electronic design automation (EDA), and Very Large Scare Integrated (VLSI) circuit design. An example of one of my other projects involves “smart” hardware verification. In this work, we use Natural Language Processing (NLP) to analyze hardware specifications and automatically generate formal properties which let us test if a design is functioning correctly. This helps us to catch bugs earlier in the design process where they are less expensive to fix, perhaps avoiding a costly recall for a chip manufacturer.

Researcher’s Background

I am currently a Presidential Diversity Postdoctoral Fellow in the Laboratory for Engineering Man/Machine Systems (LEMS) in the School of Engineering at Brown University. I completed my B.S. in Computer Engineering at the University of Alabama in Huntsville and my B.S. in Applied Mathematics at Oakwood University. My M.S. in Electrical Engineering is from the University of Notre Dame, where I specialized in solid state electronics. I received my Ph.D in Electrical and Computer Engineering from the Department of Electrical Engineering and Computer Science at the University of California, Irvine where I was affiliated with the Center for Embedded and Cyber-physical Systems. As a student, I was fortunate enough to be a recipient of the GEM fellowship, the California Regents fellowship, and the Arthur J. Schmitt Presidential fellowship. I have worked as an embedded systems engineer, digital designer, and hardware verification engineer in industries ranging from aerospace to semiconductor manufacturing. I am also a member of the ACM, IEEE, and the National Society of Black Engineers (NSBE).