So much typing...

Doing corrections day in day out is pretty demoralising... I'm always torn between having noise in the background or seeing if I can concentrate without any. I tend to land in column A. Well in any case it needs doing, and the upside is I've read a decent number of recent papers that are pretty interesting, and one that I can't believe got published. It's a bit surreal seeing a publication that purposely ignores the work of your colleagues (and me too I guess, but I don't imagine I'm very sought-after).

Anyway this paper... let's call it "On decentralized coordination for spatial task allocation and scheduling in heterogeneous teams" because that's its name, is noticeably lacking in any mention of max-sum or its ilk. Very odd given its ubiquity.

Here's one last random thing. Threw these together to back up my literature review: they're examples of belief-data overlaid on spatial maps. It's the kind of thing you can create from crowd reports of building damage and the like, and shows nicely how you could use it to make path plans for a UAV to follow in those cases. There you go.

Belief maps, in various forms

Quick NY update

Haven't had much to say recently, but I doubt there are that many readers out there desperate to hear about my every move. Short summary time, then.

Basically I've spent the last few weeks trying to get my head around ROS, a software package for coding operating protocols, algorithms, navigation, message passing etc for robot hardware. It's good for simulations too, which is what I'd like to have done in a week or so. I suffered a big-ish setback when I found out the version I was using was obsolete and had to start again with the new build environment called "catkin". I'm guessing the changes are mostly under the hood because it seems somewhat less intuitive than the first version.

So now I'm spending my time trying to work out how to get Max Sum into a form suitable for simulation, which is quite daunting given that I have no experience in software engineering before now and with multiple interacting parts to consider, that's very much what I'm faced with. The sums and functions themselves aren't all that complex, although translating my understanding of Max Sum from a simple graph-colouring exercise into task allocation is proving trickier than first imagined. Let's see how it goes.

-Chris