Several years ago, I started learning how to fly kites, as well as going to a few kite festivals, where I saw team kiteflying for the first time. Competitive kiteflying is “a thing”, and as with many acrobatic sports, there are often competition rules that include compulsory exercises. So for example, here’s how an exercise for a three-person team can be described:
One interesting thing to note about this sort of diagram is how it manages to capture time. As with XKCD-style narrative charts, multiple points of time and multiple actors are represented on the same diagram and the flow of a course of events by multiple actors is recorded via a timeline, literally, that needs to be read as such.
Of course, just how much you can fit into a single diagram is limited, so to design a complete routine requires multiple steps, or “scenes”. A good example of this can be seen in chapter 4 of Sport Kite Magic: Magic and Music: Designing a Kite Ballet [PDF].
A couple of years later, I started doing a range of robot outreach activities, which included robot dance (in part under the auspices of the RoboCupJunior robot competition) and a robot fashion show (run in association with the Quay Arts Centre on the Isle of Wight, and which culminated in a full on (ish!) theatre performance, and as such remains my favourite ever outreach activity!)
One idea we started to explore was that the robot dance outreach activities, which often took place in after-school robot clubs over a series of weeks, might be expanded to include a notation based design element based on something like Feuillet’s notation from 1700 or so, reflecting the fact that much of the robot choreography related to getting decorated mobile robots to perform synchro routines within a known stage area.
At the time, we were also looking at a variety of ways of developing maths-and-robotics activities. “A Robot’s Story”, which required kids to instrument a robot with sensors, chart the sensor readings, and then recreate a story about what the robot has seen and done, was one such activity. This later found its way, in part, into an OU residential school activity that still runs each year. Introducing geometric reasoning and calculations into a dance routine designs was also considered but never really pursued.
(Around about the same time, I’d set up (and actually got funding for) the EPSRC/AHRB Creative Robotics Research Network. One of the things I’d started to explore in that context was how arts practice could inform the documentation of robotics research activity. For example, I’d been to several robot conferences where the behaviours of robots being described were often undocumented visually, with the robots themselves being long since cannibalised for parts for other, newer robots. It seemed that perhaps the network could help, by putting robot practitioners in touch with sculpters, environmental artists and performance artists in learn from the arts about effective ways of documenting ephemeral or movement based performances. For all the successes of the network, I don’t think we pursued that aspect as far as a deliverable…)
Having a quick look around the web more recently, ways of describing compulsory ice-skating routines look like they might have been relevant too (and these don’t totally dissimilar to Feuillet notation either?)…
Where we want to program one or more mobile robots to performa different actions at different stations, perhaps we might take guidance from the specification of compulsory aerobics gymnastic routines?
As well as using simple mobile robots for kids’ robot dance routines, we also got a couple of Robonova humanoid robots that stood a couple of feet high.
We never used this for a dance routine of our own, but we did use them for demos. As well as being programmable, the robots had preprogrammed behaviours that could be triggered from a remote control handset. Some of the behaviours were quite elaborate. As well as holding it’s arms out the robot could do a squat and stand up again, stand on one leg, do a handstand, and even do a cartwheel. It used to be quite amusing demoing the robot in primary schools and getting the kids to try to mimic the behaviours, getting them to think about how they had to bend and which joints they had to bend, as well as the muscles associated with making a particular bend, compared with the robot’s single axis servos, though we tended to avoid getting them to try the handstand and the cartwheel!
Thinking about it, we never considered notations for choreographing the Robonova’s activities, though again, the world of dance provides several examples we might have been able to draw on, such as the Benesh Movement Notation:
We’ve just invested in several Baxter paraplegic humanoid robots and I wonder if we might try to develop, or better, reuse, some sort of notation – tied to preprogrammed movements – for choreographing some moves for it..?
(Finally, just in passing, I also recall programming a hexapod robot (the robot was about 2 feet long) to do a simple repetitive set of movements within a circular envelope (forward, back, turn a bit, forward, back, turn a bit, etc) for an OU Open Day (when we still had such things). One thing that was particularly interesting about this floor demo was that people ended up standing just outside the envelope/effective perimeter of the robots exploration area, so it looked as if the robot was walking up to someone, detecting them, then reverse, turn and move forward again. It would have been interesting to see if we could get audiences to arrange themselves (unwittingly) around different envelopes/action area, i.e. whether we could effectively programme the emergent positioning audience though the robot’s (fixed) behaviour, but we never tried… I wonder if we could do the same with the Baxter, or whether folk would naturally stand just more than an arm’s length away?)
PS Possible related: @pegleggen/Genevieve Smith-Nunes, DataDriveDance (/ish via @eyebeams).