Hi Samarth,
yes, I agree with your analysis. One aspect I'd like to emphasize,
though, is that with 'intelligence' we need distribution, i.e., the
sensors need to talk to each other.
Now for the discussion on the use case: You're perfectly right, if we
predict we are operating under probabilities. First assumption is that
the person continues moving and does not simply stop.
Sometimes these probabilities are very high, for example, when a person
walks down a hallway or opens a door. Sometimes these probabilities are
lower, if a person can walk right or left for instance.
Our use case is that we want (with some threshold) support continuous
lightening on some price of energy consumption. Typically light sources
(incl. the sensors) are not densely covering the area. In our university
for example, I have to deeply enter a lecture hall until a sensor
detects me and switches on the light. This is really unpleasant at
complete darkness. It is also inconvenient, if people walk down a path
and lights turn on too late. Then a steady change in brightness is
affecting the vision.
However, there is some tunable parameter (energy versus comfort) and
there is learning potential (i.e., how often did a prediction succeed?).
I guess this gives a rather rich field of building a solution that
covers quite a number of aspects and focuses on the really exciting
topic of a distributed intelligent system.
Does this help?
Best regards,
Thomas
On 09.03.2015 09:03, Samarth Bansal wrote:
Hello!
I just got across Project A2 : Intelligently interacting light switches.
Sounds really interesting.
I have a few questions:
1. The most naive approach I can think about it is this - When a motion
detector sensor detects the presence of a person, the light turns on.
2. Now that we are adding intelligence - is it that basically we are
trying to determine the path that a person might take and then turn on
lights accordingly? Lets suppose that the particular person has an ID,
and our devices have historical data about that person as well as of
other people. So the path is determined by taking into context and
giving apt weight to both forms of data - to predict a path.
3. If that is the case, I have a concern. As for any learning problem,
we will be operating under probabilities. Given the nature of the
problem in hand, there maybe multiple paths possible with high
probability(Right?). For energy efficiency, we won't really like the
lighting system to turn on all the probable lights. That is the precise
reasons for having sensors. So how does the intelligent lighting system
help?
The hardware part looks simple. I guess that any micro-controller with a
PIR motion sensor attachment and a bunch of LEDs should be good enough
for prototyping. Although I think we can simulate the environment by
laying down a graph of motion sensors and lights instead of setting up
hardware. Does this make sense?
Looking forward to the comments!
Thanks,
Samarth
--
Samarth Bansal
4th Year Undergraduate
Department of Mathematics and Statistics
IIT Kanpur
E-Mail: samar...@iitk.ac.in mailto:samar...@iitk.ac.in,
samarthbansa...@gmail.com mailto:samarthbansa...@gmail.com
Phone: +91-9871551169
Blog : samarthbansal.com/blog http://samarthbansal.com/blog
--
Prof. Dr. Thomas C. Schmidt
° Hamburg University of Applied Sciences Berliner Tor 7 °
° Dept. Informatik, Internet Technologies Group20099 Hamburg, Germany °
° http://www.haw-hamburg.de/inet Fon: +49-40-42875-8452 °
° http://www.informatik.haw-hamburg.de/~schmidtFax: +49-40-42875-8409 °
___
devel mailing list
devel@riot-os.org
http://lists.riot-os.org/mailman/listinfo/devel
