Jump to content
Sign in to follow this  
Mr.RC-Cam

BTA Autopilot

Recommended Posts

The BTA Autopilot ads claim it is a new design. I think the latest change is the addition of the optional GPS return home feature, but I'm not sure. Has anyone tried this product lately?

http://www.maxxprod.com/mpi/mpi-16.html

BTW, I looked at the patent and I am a bit confused on the performance of the technolgy. It appears to be basically an altimeter/barometer to control the elevator servo and a gyro to handle the aileron servos. I can see how this would work if the autopilot took over while the aircraft was fairly level. But if the model was in a roll when the autopilot switched on, it seems to me it can't recover. Can anyone offer some real world experience with this scenario?

Share this post


Link to post
Share on other sites

The BTA AS-06 and now the BTA AS-07.

I own, and have flown each of them. There's a little more to it. There are no mercury switches in these devices. Since the 06 is obsolete we'll just talk about the 07. The biggest difference is that the 06 used a mechanical gyro and the 07 has upgraded electronic processing and uses a piezo gyro. The 06 consumed 100ma to operate and the 07 has cut that down to 30ma. Reaction in flight seems the same to me in my 10 foot wing span 20 pound plane. (Your mileage may vary) my plane is set up to stay upright as a photo platform. (I'm not aware that there ever was a heli version. That may have been a similar product from a different manufacturer.)

The GPS function you refer to is not yet employed. The manufacturer calls it a "pigeon home" function. The 07 comes with a port to connect it to, but the GPS device hasn't been released yet. It will likely be several hundred more dollars and will only return your plane to where it last powered up at. It won't be useable to fly anywhere else. AMA should allow this device when fully employed at AMA sanctioned fields and events. Like any other electronic device in our industry, If properly set up, it will do as advertised. This device coupled with some type of failsafe function would most likely always save your plane once the GPS device is released. (Naturally if your going 100 mph, 3 feet off the deck, Inverted, down the runway center line and you lose radio contact, all bets are off.) :o

Roll control is done by a rate gyro. I don't know what king of oogedy boogedy magic they used in the ic chips to make it work. It WILL return your plane to level flight. The gyro is set inside to detect yaw changes. Yes, Yaw. It does work very well though. You can see it on the ground as you taxi your plane. When you start to turn (Yaw) you will see the ailerons deflect. In the air the combination of Rate gyro watching yaw, and whatever they have those chips doing will put your plane flat in roll in very quick time. It will return it to upright if your inverted. (The time it takes depends on your planes response characteristics and what attitude your in when you employ it) You can fly with it fully employed. It will just give you less roll response and if you release the stick it flattens right back out. Typical response time is a second or less.

Pitch is as you stated. Barometric sensing. It won't hold your altitude forever if left employed. It develops a reference voltage from the barometric sensor and then works the elevator up or down to keep the voltage the same. (Pressure from atmosphere) What it will do is keep your plane from climbing or dropping when employed. If your in a screaming dive, and you enable the device from a spare channel, it will pull you right out. I observe the plane climb a little after it pulls out. I'm certain the device is trying to get to where it was set at when employed. Over a long range or in high winds, it will rise and sink a little. You can move your elevator when it is engaged and still change pitch. It will try to keep your plane where it was last left at. The pitch also hunts a little. The barometric sensor technology isn't super sensitive. It might porpoise in a 20 foot or so band of altitude. There are gain adjustments on the unit that allow it to be set up best for your own planes response characteristics. Thats for both pitch and roll. If it porpoises noticably, there is probably too much gain set to the servo and it's over responding. You can have a plane full of holes to get it to work well either. Vented is OK. Air raming in old fuel line holes or oversized control rod holes in the firewall have to be plugged. You can't have the plane experiencing air pressure changes internally for the altitude part to work well.

It operates and reacts a little differently than the FMA co-pilot that I also fly. Each has it's pro's and cons and I am happy with both units in their own specialty areas.

Dan

Edited by kd7ost

Share this post


Link to post
Share on other sites

Thanks for the details. It is still a mystery to me how it can return to level flight if the Autopilot was enabled during a significant roll. I'll have to dig deeper into the patent info.

Share this post


Link to post
Share on other sites

I'd be curious to know how it's done as well. I've tried standard gyros set up for roll control. Nothing works. Even trying one set to correct roll when set up on yaw axis like the BTA unit. Whatever makes it work is how they process the error signal from the gyro to make it level. The original BTA AS-06 was developed by the Isreali's for UAV flight. I think it's still for sale in Isreal.

http://www.jandgpecans.co.il/index.html

Maybe the 07 is a knock off sold by maxxproducts? They're the ones who imported the 06. Just speculating out loud. I forgot to mention the 07 is also smaller but they use the identical wiring harness.

Dan

Share this post


Link to post
Share on other sites
Roll control is done by a rate gyro. I don't know what king of oogedy boogedy magic they used in the ic chips to make it work.

Thanks for the detail review. Do you know if there is an accelerometer chip on the board?

Share this post


Link to post
Share on other sites
Do you know if there is an accelerometer chip on the board?

Yes, that is the missing puzzle piece. If it has one, then it would offer the traditional approach to stabilization via a kalman filter. However, it is not shown on the patent filing. I didn't see tilt switches either.

I read the patent again and I think I understand how it proposes to operate. The key is the barometer sensor.

The goal is to keep the model at the current altitude. So, when the autopilot takes over, it uses the barometer to determine if the model is climbing or falling. It will then use this feedback, as well as the feedback from the gyro, to manipulate the elevator and ailerons until the altitude is stable and is no longer rolling. Regardless of actual model orientation, once the two sensors are satisfied, the model will be held in that path. A stable model would easily self-right with its help.

I have a feeling that it would be a wild azz looking flight path if the model was upside down or in a steep bank when the auto pilot was enabled. With enough altitude and a predictable model, it would be able to recover eventually. Not a glamorous way to do it, but practical. A model heli or fast unpredictable aircraft (like a high speed jet) would be a bad fit for this simple approach.

Share this post


Link to post
Share on other sites

Dave is correct. The co-pilot also works one heck of a lot faster in pitch. The sensor technology used in the co-pilot balances the horizon all around. Just the slightest pitch or roll off shows up as a difference in the sensors. It keeps things real solid. You don't see the slight hunting in a barametric sensing setup. If your just flying RC at your local field, you shouldn't see it hunt if you have the right gain dialed in.

The only times I find the co-pilot needs to be swapped out is along the Snake river for example. There are places where cliff's on one or both sides interfere and the device won't work to keep the plane level. Also when it gets winter time up here and it's real cold and cloudy. The co-pilot needs to see the IR temperature difference. If it can't se the difference it won't roll control. Times like that is when the BTA AS-07 pulls me through.

Dan

Edited by kd7ost

Share this post


Link to post
Share on other sites

Cool Dave,

I have all your video's downloaded on my PC at work actually. I don't know where you were at when that flight was filmed. Thats neat to know though. Were the other flights up here in Idaho all flown with the co-pilot?

There's a place just South of Nampa where I live where one side is a huge cliff and the other side opens up to the Owyhee Front. I tried to film the river there and the plane kept the cliff side wing way up high. I was pretty confident with my calibration though. I flew that flight without pitch control employed in the co-pilot. So I just aimed the planes nose at the wall and calibrated for roll balancing out open area on both sides.

Anyway the flying was wierd. The plane kept the cliffside wing high like I said, so it kept slipping to the downside wing. The elevator kept going up with the PDC-20 trying to keep alttude. The plane just kept turning away from the river no matter which direction I flew. It flew flat once it was going straight away or when I had it flying straight at the cliff. I spent the afternoon screwing around with trying to re-cal different ways but nothing worked. Thats where I got the idea you need a good horizon view, or at least, like you mentioned above, equal on both sides.

I dunno, you think I might have over looked something? You've been doing this a lot longer than I have.

Dan

Share this post


Link to post
Share on other sites

RC-cam,

I'm sure you've checked this out but just in case you haven't.... www.carvec.co.uk

Sounds like your looking to roll your own I suspect.

Dan

Share this post


Link to post
Share on other sites
Sounds like your looking to roll your own I suspect.

Actually, I was just curious how it worked. That is all.

Share this post


Link to post
Share on other sites

My hats off to all the software geeks that make the hardware work. I consider myself pretty skilled in a systems arena. I'm a very instinctive fixed wing pilot that will take off and put big planes down on little bits of two track with sage brush just off the wing tips. The parts I buy to give me autonomous or at least stabilized long distance flight blows my mind as to how well it works. I work to understand those flight system devices. That having been said, I can't immagine what the rotor wing guys go through. Those things shouldn't fly to begin with. My buddy calls them a million seperate parts all flying together in close formation. How you make a chopper sit still, and fly by GPS when you have that whole going backwards thing going on is a bigger mystery to me than how the BTA AS-07 levels wings in roll through a rate gyro watching yaw. You Helicopter guys have bigger problems to solve than I can immagine.

Good luck with your ventures. I'm gonna go fly in the snow.

Dan

Share this post


Link to post
Share on other sites

I think I am right in saying that the co-pilot dose not hold height, it only holds trim. If the motor stops what happens ? I assume it will stall.

Terry

Share this post


Link to post
Share on other sites
So, when the autopilot takes over, it uses the barometer to determine if the model is climbing or falling. It will then use this feedback, as well as the feedback from the gyro, to manipulate the elevator and ailerons until the altitude is stable and is no longer rolling.

That makes sense to me. In an airplane a non zero roll will either create a descent (if elevator is not adjusted to compensate) or a yaw change (as during coordinated turn with positive elevator). Clever!

I don't see how this can be applied to a helicopter though, unfortunately.

Share this post


Link to post
Share on other sites
I think I am right in saying that the co-pilot dose not hold height, it only holds trim. If the motor stops what happens ? I assume it will stall.

You are correct. Even more so though, the plane will not hold altitude with the engine even running at full speed. The device keep the plane flat in roll and pitch. It won't stop it from rising or sinking with normal lift and air characteristics.

Dan

Share this post


Link to post
Share on other sites
How then can it put this to a horizontal attitude if it thinks it's already level?

If my assumptions are correct, it is not really trying to put the model horizontal. Instead, it is trying to find a way to maintain altitude. Doing so, by a set of elevator and aileron movements, will result in a model that will eventually seek a horizontal-like attitude.

There are cheap mechanical toys that run on the ground. They are designed to get themselves upright when they flip over. Technically, the BTA does not work like that of course. But, in a conceptual sense, they seem {to me} to be similar.

Share this post


Link to post
Share on other sites

Any bank angle will result in a turn as the altimeter tries to kill the decent by pulling in elevator so this will cause the aircraft to turn. This then is sensed by the gyro and will level the aircraft. For this to work the gyro needs to be very sensitve. I have used a gyro for this with no altimeter function to help and as long as the aircraft is trimmed ok it works fine. I tried using a PG03 piezo gyro but it allowed the aircraft to circle allthough it did prevent a spiral dive. The best I have found is to fit very weak springs to a michanical gyro

Share this post


Link to post
Share on other sites

It will help to figure out what BTA is doing by making a simple test on the ground. While holding the plane:

1. Roll the plane and see if ailerons get deflected by BTA.

2. Yaw the plane and see the aileron response.

You should see aileron deflection only in test #2. If ailerons move in test #1, there is something else inside BTA in addition to what we discussed here, like accelerometer or roll gyro.

Share this post


Link to post
Share on other sites

Here is a snippet from the patent that summarizes the trickery:

In the present invention, however, by sensing rate-of-turn, a signal proportional to the actual bank-angle is produced; and by sensing rate-of-climb, a signal proportional to the actual pitch-angle is produced; these latter signals are applied, with the bank-angle and pitch-angle command signals to the aileron and elevator drives, respectively, and constitute negative feedback signals tending to stabilize the aircraft with respect to bank-angle and pitch-angle.

For example, when the bank-angle command signal is given, the rate-of-turn sensor on the aircraft produces an output signal which opposes the bank-angle command signal, such that in the steady state, a constant bank-angle is achieved; this produces the commanded turn of the aircraft while the bank-angle is stabilized. Pitch-angle stabilization is produced in the same manner by the negative feedback signal generated by the rate-of-climb sensor, which measures the actual pitch-angle and not rate-of-change of pitch; this signal is applied to the aircraft elevator drive with the pitch-angle command signal and thereby produces the commanded climb of the aircraft while the pitch-angle is stablilized.

The patent also claims that the gyro (referred to as a ROT in the filing) senses roll rather than yaw.

Share this post


Link to post
Share on other sites

I mount my gyro at 45deg so it senses roll and yaw, it sounds the same as in their test.

Terry

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×