[Ns-developers] Updates on GSOC Uan Framework
andrea.sacco85 at gmail.com
Tue Jun 1 07:43:39 PDT 2010
I would like to let you know my progress for the gsoc project.
I've done an overall design of the classes I'm going to implement:
- UanAuvMobilityModel, inherits from MobilityModel
- UanRemusMobilityModel, inherits from UanAuvMobilityModel
- UanGliderMobilityModel, inherits from UanAuvMobilityModel
I've asked the producers of the two AUV to know the operational
characteristics, like max depth, max pitch, max speed, etc.
I've started implementing the Remus model, as from project timeschedule.
I've some questions:
- Is an useful abstraction the UanAuvMobilityModel or it is better to derive
the two concrete AUV classes from MobilityModel?
- I've noticed some approximation problems, using the constant velocity
model. I've implemented two methods, Emerge and Submerge, to let the vehicle
emerge/submerge to a certain depth. Computing the emerging/submerging time
with vertical velocity and target depth, and getting the position on this
time, it returns me a value with an error of about 10^-9 (m). How can I
solve this problem? Should I round the position coordinates?
Regarding the Energy section of the project, I will implement the following
- MicroModemEnergyModel, it should inherits from RadioEnergyModel but, I
find it could be more generic to rename the interface to
TransducerEnergyModel, as a transducer model could be a radio but also of
another type, acoustic in this case.
- LiIonEnergySource, inherits from EnergySource. I think we should talk
about how to implement a non-linear discharge. I've thought to implement a
"fitting" function for the V/A discharge curve, but I hope to hear something
Waiting for your comments.
More information about the Ns-developers