The Sikorsky X2 Test Flight

It looks as though Sikorsky is on the path to creating one of the fastest helicopters in the world. Their new X2 helicopter just recently had its first test flight using the high speed pusher prop. The latest flight test didn’t break any speed records at 40 knots, but Sikorsky is planning for 250 knots by the time all tests are done. The current record holder for fastest helicopter belongs to the Westland Lynx which set a world record speed of 216 knots.

The Lynx HAS3 similar to the ZB500

The X2 helicopter looks pretty cool. in a traditional single rotor helicopter, the retreating blades at high forward velocity start to lose their effectiveness. The idea behind using the the coaxial rotors is that you will have both rotors counter-rotating which will provide you with an advancing blade on each side of the helicopter. This fixes the problem of the aircraft’s stability. The pusher prop at the back gives the X2 the extra thrust to be able to get the helicopter past the speed record.

A few more of the specifics of the Sikorsky X2 can be found at its wikipedia page.

I created the driver for the autopilot to be able to read from the new laser altimeter.  We made some small changes to the autopilot in order to take advantage of the faster update rate and setup a terrain following option in the operator interface.  This took care of setting up the software side of things.  A helicopter mount was created and the new laser was attached to one of the SiCX 12 UAVs we have.  All that was left to do was to perform a flight test.  We performed a few flight tests to verify the accuracy of the laser altimeter and also the commanded response time.  These tests were done to verify that we weren’t going to run into the ground when we started performing terrain following.  Everything looked good.  We recently were able to flight test this new feature and found that it performs very well.  The video, at the bottom of this post, was taken of the SiCX 12 UAV performing terrain following at 6 feet above the ground with a forward velocity of about 5 knots.  We could watch the helicopter go over small hills, climbing and descending with the terrain. Pretty cool!

Terrain following is a useful feature, you may not always have prior knowledge of the elevation of the terrain you are flying over.  You could always just fly high and not have to worry about hills, but sometimes you need to fly close to the ground.  If for instance you want to be able to fly underneath certain objects to get to a location, or if you have close range sensors that you are using to sense different items on the ground in a search and rescue mission.  These type of applications require that you stay a specific distance away from the ground, while not running into the ground, which is what this sensor has provided for our autopilot.

The North Little Rock Police Department in Arkansas has been testing out a helicopter UAV to add to their police force. It is was planned to carry out barricade and crowd monitoring for large events according to the Argenta News article. They just recently suffered a crash due to software problems according to the latest news article.
The news article refers to the UAV as a toy and raises questions about the money being spent towards the UAV. However, the news article fails to mention the costs of operating manned helicopters including the risks involved with the pilot’s life. UAVs are far from toys. They have been proven to be very beneficial in many different applications without risking a pilots life while also saving a lot of money. As a cost comparison, the Dekalb County Police Department in Georgia uses two Eurocopter AS-350 B-2 helicopters in their fleet. These are manned helicopters that require a crew of two people. New, these helicopters go for $1.6M and cost approximately $350 an hour to operate plus fuel and crew’s time. With a UAV you have much much lower operating costs, less fuel, less training, less time spent getting in the air, and less of a risk to the crew.

To me UAVs just make sense. I guess not everyone sees that.

The following video shows different prototypes performing various maneuvers.  It is pretty incredible the size and weight of this vehicle.  It will be interesting to see how they incorporate collision avoidance and a navigation system into such a small platform.

Linear controllers have often been used to control autonomous aircraft. These complex vehicles often require aircraft dynamic models that are linearized about a given flight condition, such as a hover. However, aircraft dynamics are nonlinear therefore with a linear controller you have to spend a significant amount of time developing an accurate aircraft model of the vehicle dynamics across the entire flight envelope, or you spend time playing with and tweaking gains until you get a controller that is ‘good enough’. This gets really difficult when you have a complex aircraft transitioning from one flight condition to another, such as a VTOL (Vertical Take-Off and Landing) fixed-wing vehicle transitioning from a hover to forward flight. The aircraft dynamics during this transition are complex and are hard to model.
At my work we use a controller that is called an adaptive neural network controller. It is specifically designed to adapt to aircraft changes, such as in flight failures or payload changes. It eliminates most of the aircraft modeling and tuning that you typically have with linear controllers. The keyword I use to describe it is ‘flexible’ and it is able to be used on a wide range of autonomous aerial vehicles with very little changes in the controller. This means that when a new complex vehicle is designed, the controller can be put on the vehicle and fly it without many changes being done to the controller.

I went on about the control theory a little more than I thought I would. Oh well, it was fun trying to explain it. I definitely enjoy working at GST as I get to learn more about autonomous helicopters. Being able to work on the software to help control these vehicles, and then going out to the field and watching them fly is really cool!  I will end this post of a picture of one of the UAV helicopters I work on, called the SiCX 12.

SiCX12 Helicopter UAV