Hello all,

I'm working on a robot-catamaran. Its role is to make measures of the bottom depth on several points, in order to create bathymetric maps of lakes or ocean. I attached a little pict of the catamaran on its way for a survey. 

At this point it is able to follow a route by itself quiet well, but it is sometimes going away from the route a little bit too much. 

To solve this I have tried to implement a "Cross Track error" (XTE is navigation, sometimes it is also called XTK if I'm right) correction algorithm to make the boat get closer to the route.

But... It's not so good. Cf the second pict. It's oscillating. 

Maybe the algorithm is bad, maybe its settings are bad, I still don't know, but I feel I can spend hours on this while it is probably a well-known and well-solved problem!

SO : 

Is there an existing algorithm / formulae to do this?

Note : A friend of me gave me the following formula, but I don't know how to use it exactly. It is :

Steering signal = 0,0281XTK + 0,002 x GS x TKE

One of my problem, beyond the formula, is that my microcontroller "gets" the position with 2 sec of delay. It partly explains why the boats turns so far away after reaching the waypoint. I'm using a GPS EM406, and I read NMEA. And all this is slow!! 

How do you people handle this problem of latency? Or even, do you encounter this problem of latency (maybe it is a bad architecture in my code)?

Thanks for your answer(s)!

Mathieu

DSC03591.JPG

Route and Track.JPG

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Did you solve this ?   I have implemented an AP for my 34' boat that has steered me to Hawaii and back and can explain how I handle this type of issue.

    Brian

This reply was deleted.

Activity