A Framework for Realizing Extensible Sensor Platforms

Gavin Strunk
Seminar

Embedded computing has improved the way we collect observations, control our environment, and operate systems. We believe that there is a shortage of sensing platforms that combine the abilities to interface and interact with a wide range on sensors ranging from a simple temperature sensor to a more advanced Infrared Imager. The diverse set of sensors and analyzers that we wish to gather data from, interface to computing hardware through several communication protocols including Rs-232, I2C, USB and TCP/IP. There is consequently a need for appropriate hardware and software support on the sensing platform. In an effort to support the various sensors and analyzers, we have designed and realized a sensing platform using the BeagleBone Black embedded hardware modules. In this talk, we will start with a high level overview of the developed hardware architecture. Next, we will be address the numerous difficulties associated with implementing such a system, from both conceptual and practical perspectives. I will follow with a demonstration of the necessary steps to implement a new hardware system to fit within the designed architecture. I will wrap up the talk discussing how a variety of data related sensing problems can be solved more efficiently using the system, cloud computing couplings, and our future directions.

Bio:
Gavin Strunk is a summer research aide working in the MCS Division. He is pursuing a Ph.D from the University of Kansas Mechanical Engineering Department with a emphasis in intelligent control systems. Gavin’s background involves the application of artificial intelligence and embedded controls to complex hardware systems. His current research focuses on developing an intelligent optimal controller and research grade data acquisition system for an ultracapacitor bank as part of a mobile hybrid power source.