Instructors: Dav Clark, Javier Rosa & guests from industry, the social sector and academia
The issue currently is that it is often difficult to get accurate information in very dangerous areas. Subsequently anecdotal evidence and information from the governments are often used, which have obvious problems. We want this to be a part of a multi-modal system for remote conflict analysis. To be clear, this is to gather more information, not to replace on the group reporting, or other “ground truth” systems but to support them.
The rcSensing project is focused on using sensitive accelerometers as strong-force seismometers. The intent is to be able to identify events in conflict zones, which can very quite widely. Some of the potential events we intend to detect are.
Obviously these are not common events in Berkeley and so we will first be approximating strong force via tracking trains. We will be working on detecting the difference between freight and passenger trains and then work on detecting BART trains which should comparatively have a much lighter signature. Once we accomplish this, we plan to experiment with detecting multiple cars on dirt roads which will give us a closer proxy to an event in the field. From there, we would like to start collecting data in the field which we can then analyze but this may be outside the scope of this class.
name | github handle | nano-bio |
---|---|---|
Bryan Morgan | @brynamo | Graduate student, School of Information |
Juan Shishido | @juanshishido | Graduate student, School of Information |