Efficient Mapping and Management of Applications onto Cyber-Physical Systems
Abstract:
Cyber-Physical Systems comprise richly heterogeneous collections of devices (mobile devices, home electronics, taxis, robotic drones, etc.) that together gather sensor data, analyze it, and coordinate large-‐scale actions in response to it. While we are increasingly deploying and relying on such systems, the reality is that it remains difficult to program them well, and also difficult to manage their operations in ways that optimize data delivery at acceptable performance and energy levels. Our project attacks a vertical set of challenges related to mapping and managing CPS computations. Our most recent work includes a testbed for mobile sensors, a set of motion planning experiments for robotic sensors, as well as work on optimizing communications energy through prediction of data arrivals, data delay tolerance, and communication availability.
Submitted by Margaret Martonosi
on
Abstract:
Cyber-Physical Systems comprise richly heterogeneous collections of devices (mobile devices, home electronics, taxis, robotic drones, etc.) that together gather sensor data, analyze it, and coordinate large-‐scale actions in response to it. While we are increasingly deploying and relying on such systems, the reality is that it remains difficult to program them well, and also difficult to manage their operations in ways that optimize data delivery at acceptable performance and energy levels. Our project attacks a vertical set of challenges related to mapping and managing CPS computations. Our most recent work includes a testbed for mobile sensors, a set of motion planning experiments for robotic sensors, as well as work on optimizing communications energy through prediction of data arrivals, data delay tolerance, and communication availability.