Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 Hello Forum, the whole discussions about "muscle memory" got me onto this: How about messing around a bit with the actual medical basics underlying what we experience in FPV RC? There's a lot of "magic" information around, where it's basically just nature doing it's thang. I'll give an example: When we are trying to react to something we see motorically (e.g. preventing a quad from smashing against smtg. from a wind gust), the best reaction time we can achieve is 78 milliseconds, because the information has to go all the way from "the wires" (nerves) of our eyes to the "little visual brain" to get interpreted, then to the premotorical cortex for motion planning, and then down our spines (biological esc's :-) to coordinate muscle movement (..in a feedback PID-loop with our cerebellum, the part of the brain residing on top of our spinal cord, just after the entrance into the skull). That takes time.. In martial arts, we learn to react to tactile impulses and condition our movements based upon "polysynaptic reflex loops", meaning simply that there are a couple of loops working together independently from our brain. The information doesn't have to "turn" in our brains, but can directly unleash the pre-programmed reaction pattern. Of course we need drill exercises to do so, but it's possible even for people without brain ( ), so there should be hope.. In this case (polysynaptic loop), we manage to react within 12milliseconds, isn't that great? Another limiting factor is our "framerate", thus fps of our eyes, which is only around 25 fps - so never try playing reaction games with a bear, because - even though suffering a bit more inertia than us, being that heavy - they see the world in at least 40fps, so we'd be screwed independently from our training.. So what does this mean for FPV flying? Well, as long as we solely rely on visual information, we won't be able to further reduce reaction times. We're stuck @ 78 msec. BUT... ..if we'd manage to get some feedback over our sticks (..thus haptic feedback, "touch"-feedback), we theoretically could reduce reaction time down to 12 msec!! I don't know if that's an issue yet, but I'll keep playing with this thought a bit, in case anyone is interested. My goal is a direct quad-human-interface, one day we might just think what the quad has to do, instead of "sticking it" - even though I reeeeaallllyyy LOOOVE juggling my sticks! One doesn't exclude the other, though.. Before venturing such bold projects , I'll just try getting my 9oGT & Shuriken working. At the moment, reaction time of my equipment is up to.. ..a couple of hours? So long, keep it up, fellas, this forum is seriously great stuff, thanks again to the founder, Oscar!! Ga-reat breeding ground for future technologies & concepts! FPV-on, Emanuele Posts: 2,416 Threads: 51 Likes Received: 1,861 in 1,175 posts Likes Given: 3,315 Joined: Mar 2016 Reputation: 74 Interesting idea! Seems like getting the telemetry information back to the Tx would be the easier part (except what about latency?). The harder part would be re-designing the Tx to translate the telemetry to feedback (vibration, movement?), no? Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 (21-Mar-2017, 10:04 AM)sloscotty Wrote: Interesting idea! Seems like getting the telemetry information back to the Tx would be the easier part (except what about latency?). The harder part would be re-designing the Tx to translate the telemetry to feedback (vibration, movement?), no? Heya Scot! Glad to know someone follows my strange ideas I guess the Tx is not the problem in terms of latency, rather the elaboration protocols: We'd have to find some kind of force-feedback mechanism which functions independently from the FC, since those "regulatory" functions would be done by our body, so cycles could be kept shorter. Anyway, it's not like we'd HAVE to react to every input, there would just be this alternative channel of information/feedback adding onto what we're already doing. The fine-tuning of quad movements would become much more immediate, instinctive - we'd hardly know what we're doing since it all happens "outside" of our brains. The whole RC system would become something like an accessory organ, over time we'd stop to notice, as the nervous cells are highly dynamic in organisation. (Nobel price in medicine recently - it's called neural plasticity and describes the time within there happen STRUCTURAL changes/adaptations to our nervous cells. Cycle time was believed to be months, but in reality it's just seconds. That's mind-blowing to me..) I have little to lose since my quads are battling my RC anyway, so no harm in messing it up additionally.. • Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 ..for example an independent accelerometer-feedback-loop, or something linked to air pressure.. It wouldn't be for general control of the aircraft at first, but sort of a "biological trim"-measure, the way I see it.. All serving the purpose of "feeling" your craft instead of having to rely entirely on visual information.. "I feel you, bay-bey!!" ;-) • Posts: 2,286 Threads: 38 Likes Received: 1,527 in 995 posts Likes Given: 1,881 Joined: Apr 2016 Reputation: 72 Direct neural interfaces are the future, but (unfortunately) not a future I'm likely to experience. My children may, or possibly their children. At the moment, all the ways we have at our disposal to interface directly with the brain are far too crude and cumbersome. Posts: 2,416 Threads: 51 Likes Received: 1,861 in 1,175 posts Likes Given: 3,315 Joined: Mar 2016 Reputation: 74 21-Mar-2017, 11:27 AM (This post was last modified: 21-Mar-2017, 11:31 AM by sloscotty.) By latency, I just meant the latency between the FC and the Tx. Getting the required current flight config/attitude information processed (probably a lot) on the FC, and then on to the Rx via serial channel, and then back to the TX will take "time". Can that be low enough to provide an advantage over what you see through the goggles? (And that's assuming the translation in the Tx to haptic feedback is already taken care of.) It's kind of the hardware aspects of this that interests me most. I'm a tinkerer at heart (although this is WAY beyond my skills). Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 (21-Mar-2017, 11:21 AM)unseen Wrote: Direct neural interfaces are the future, but (unfortunately) not a future I'm likely to experience. My children may, or possibly their children. At the moment, all the ways we have at our disposal to interface directly with the brain are far too crude and cumbersome. Never say never, Dear.. I'm collaborating with the neuroscience-lab of the university of zürich (prof. schwarz) and they're already doing some quite unbelievable stuff! Furthermore, progress accellerates itself - it's an exponential process, and findings usually aren't planned for. (..like most nobel prizes were discoveries by chance, e.g. high-temperature superconductors, which lingered around in labs worldwide way before their discovery, but noone thought about measuring the conductivity of ceramics. "They're isolators, right?" Oh-so-wrong.. Let's start gingerly. I'm optimistic we'll all get a share of these marvels, if we're open to it. I'll keep you & your family posted.. Viva la famiglia! Happy day, Emanuele Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 (21-Mar-2017, 11:27 AM)sloscotty Wrote: By latency, I just meant the latency between the FC and the Tx. Getting the required current flight config/attitude information processed (probably a lot) on the FC, and then on to the Rx via serial channel, and then back to the TX will take "time". Can that be low enough to provide an advantage over what you see through the goggles? (And that's assuming the translation in the Tx to haptic feedback is already taken care of.) It's kind of the hardware aspects of this that interests me most. I'm a tinkerer at heart (although this is WAY beyond my skills). We'd have to add some processing device, I agree! - I'm not sure it requires "talk" to the FC, though. In my imagination, the whole computing may take place within the RC Tx: For instance a virtual model calculated from the given (integrative) comands, corrected through differential analysis, could determine a correction variable for the outputs to the FC/Rx. Sort of dynamic P settings, taking into consideration human reaction rather than solely environmental factors. It's just a fuzzy concept of an idea in my head, but if we never dare, we just get stuck in front of netflix and grow fat. I don't like the telly. I prefer getting out & dirty.. However, I'd like to express my gratitude & enthusiasm to the whole forum, and especially Oscar. I've been and still am traveling the world and witnessing lots of collectives and cultures, but this intofpv-forum is UNIQUE! I wish that the forum attitude would translate even just a fraction into the academic world, were unluckily narcissism is ruling. I firmly believe in the huge potential of the field. All of you guys are probably not entirely aware of how much of a role-model this community is to me - and I think it should be for others! One thing is certain: I'll take the liberty of telling medical professionals all over the world about intofpv, and how organically humanistic your approach is. It's the attitude we need to save society - may sound exaggerated, but that's the way I feel about it.. Kowabunga! • Posts: 2,416 Threads: 51 Likes Received: 1,861 in 1,175 posts Likes Given: 3,315 Joined: Mar 2016 Reputation: 74 For it to work, at least the raw signal HAS to be generated on the quad itself (and modern FC's might have the horsepower to process before sending). Like I said before, I think that would probably be the "easier" part. Translating that to feedback on the Tx might be the big problem. If you want to "feel" the gust of wind that knocked you around, you've got to have some sort of force relayed to the sticks - that's a redesign of the Tx mechanism (servos, multi-point vibration, something...). Posts: 41 Threads: 7 Likes Received: 17 in 7 posts Likes Given: 13 Joined: Feb 2017 Reputation: 1 21-Mar-2017, 03:57 PM (This post was last modified: 21-Mar-2017, 04:21 PM by EchoBravo.) (21-Mar-2017, 12:04 PM)campagnium Wrote: Never say never, Dear.. I'm collaborating with the neuroscience-lab of the university of zürich (prof. schwarz) and they're already doing some quite unbelievable stuff! Furthermore, progress accellerates itself - it's an exponential process, and findings usually aren't planned for. (..like most nobel prizes were discoveries by chance, e.g. high-temperature superconductors, which lingered around in labs worldwide way before their discovery, but noone thought about measuring the conductivity of ceramics. "They're isolators, right?" Oh-so-wrong.. Let's start gingerly. I'm optimistic we'll all get a share of these marvels, if we're open to it. I'll keep you & your family posted.. Viva la famiglia! Happy day, Emanuele Sounds like we need to talk! You just mentioned several of my life goals in one shot. Human-Computer Interface, Zurich, and BCI Drones. I was looking for somewhere I could do some school work on the topic and couldn't find it. Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 (21-Mar-2017, 12:45 PM)sloscotty Wrote: For it to work, at least the raw signal HAS to be generated on the quad itself (and modern FC's might have the horsepower to process before sending). Like I said before, I think that would probably be the "easier" part. Translating that to feedback on the Tx might be the big problem. If you want to "feel" the gust of wind that knocked you around, you've got to have some sort of force relayed to the sticks - that's a redesign of the Tx mechanism (servos, multi-point vibration, something...). You're definitely the practical guy! I'm just messing around with half-cooked ideas.. On the other hand, our "classic" RC-layout has been around for quite a while, stocked up not unlike the worldwide-befeared windows OS As there are innumerable experts who know sooo much more than me about it, I don't dare get too specific on this RF technology, but in terms of field induction/hall effect, ENG-pulse synchronisation etc. (electroneurographic), aaaand now it gets a little sci-fi: phase-interference photography, free electron laser etc., these are technologies which are just being discovered, offering unimaginable possibilities.. (see through skin&muscles without ionizing radiation, free choice of quantum energy etc...) I admit I'm a supernerd as far as this is concerned. My car, up to very recently, would barely start in the morning (don't have one now ), but I've already owned three different CBCT and one micro-ct system.. If it would exist (..and I had a shielded room + the space necessary), I would have installed a micro-MRI in my practice already long ago. I'm not lacking nerd-toys, though, don't worry.. Making a long story short, somehow I think that we have to find alternative paths to our solution. A bit like ceramics as superconductors.. Okay, back to real life's (rewarding!) basics: THANK YOU EVERYBODY ON THIS FORUM for the support, you've motivated me enormously and I'm facing my future quad-struggles under a different point of view. Very keen to read the post on my 90GT-and-Shuriken-patients, will let you know.. Throttle up! Emanuele • Posts: 227 Threads: 35 Likes Received: 202 in 116 posts Likes Given: 278 Joined: Mar 2017 Reputation: 5 (21-Mar-2017, 03:57 PM)EchoBravo Wrote: Sounds like we need to talk! You just mentioned several of my life goals in one shot. Human-Computer Interface, Zurich, and BCI Drones. I was looking for somewhere I could do some school work on the topic and couldn't find it. Heya EB, ..sure! - I'm actually happy to find people "nerdy enough" to follow my deviations.. I'm sort of a sociopath (*kidding.., I'm, let's say, functional..) who never felt at ease going out in the evening like everybody else did, but always tended to accumulate all sorts of strange things, chemicals etc. Let me know what your background is and what you'd like to do. As far as Switzerland is concerned, I'm pretty mobile in the academic world (..as the "enfant terrible" of it ..), so I can link you up to whomever you want - ETH Zürich, EPFL Lausanne, PSI, Computing centre Manno, ABB research centre Baden - you name it. Won't be able to tell you where to go out in the evening, though.. ;-) Cheers & c ya! Emanuele • Posts: 2,410 Threads: 136 Likes Received: 1,790 in 1,052 posts Likes Given: 3,302 Joined: Jan 2017 Reputation: 50 Deep! |