Gesture Controlled Unmanned Aerial Vehicle
An Unmanned Aerial Vehicle (UAV) is a small to medium sized flying machine that is controlled remotely by an operator on the ground using a hand-held radio transmitter.
Drones nowadays are widely used around the world for a variety of purposes including aerial videography, photography, surveillance etc, which require skilled pilot to perform these tasks which proves to be exorbitant.
A simple gesture controller can make the task of piloting much easier. Gesture refers to any bodily motion or state particularly any hand motion or face motion. Gesture recognition has more accuracy, high stability and faster response times and will help to make drone flying easier by decreasing response times and giving a more intuitive feel, thereby decreasing the barrier between humans and machines.
Problem Statement: Controlling of UAVs using conventional methods of radio transmittance requires skilled personnel.
Significance: Introducing a gesture-based control for tasks such as surveillance, reconnaissance, recreational or commercial makes it easier to control and also doesn’t require skilled personnel which brings down the cost.
Approach: Addition of gesture recognition devices and algorithms to form an effective interface between humans and machines.
Expected Results: Gesture based controls will ease the operations and give faster response times compared to conventional approach.
a) Leap Motion Sensor
The first choice of gesture recognition by any one venturing into image processing will be Leap motion controller because of its availability and low cost. This method has been used to control and manipulate the Parrot AR drone.
b) Kinect Motion sensor
Kinect motion sensor is a motion detector and depth detection camera that can recognize gestures with the help of its two onboard cameras, IR projector and a microphone.
This approach allows the user to control a quadcopter (a type of UAV) using hand gestures. The operator will make these gestures in front of a Microsoft Kinect device, which can sense both coloured light and depth of field. Kinect is used to define more complex gestures that take advantage of the distance of the operator’s hands from the camera. Also, they may eliminate the use of motion sensing coloured gloves. The gesture recognition component is composed of a five-step pipeline shown in figure.
c) Electromyography (EMG)
Electromyography is a method to read, intercept and decrypt the electrical impulses being fired at the tip of motor nerves that gives signal to the muscles of limbs to move or actuate. Another type of this is the Surface Electromyography (sEMG) that detects signal at the skeletal muscles. This electrical signal can be harnessed to perform gestural activities.
Galileo Bionic Hand is a DIY kit at a low cost for amputees who can get it from local manufacturers that are aimed to be released in the market with the sole purpose of reaching out to people. A similar device is Hero Arm.
d) Inertial Measurement Unit (IMU):
Inertial measurement unit comprises of accelerometer, gyroscope and magnetometer that in combination gives the position, orientation and the velocity/acceleration at which a particular object is moving around in 3D space. Using these sensors for gesture control implies that they have to be mounted physically onto a glove which can then be worn and used for sending gesture signals. These are sometimes coupled with a flex sensor to bring about the actuation of motors to move joints or linkages.
In one such IMU based project, ADXL335 as the accelerometer is used and this project describes the controlling of servomotor(Futabas3003) consistent with motions of accelerometer and the 3-D visualization of accelerometer by using interfacing Arduino (ATMEGA 328P) with the processing software program can be viewed and MATLAB is used for graphical results of accelerometer motions.
With the help of wearable devices like a smartglass and a smartwatch, users are able to perceive the user’s arm movements and body movements accurately, including the moving distance and direction, by using accelerometers and gyroscopes. They have used a smartphone as the server, which combines and compiles the processed data from the smartwatch and the smartglass, and sends the commands to the remote controller wirelessly to manipulate the UAV.
Gesture controlled robotics is an upcoming fad to make operation easier and convenient and also to decrease the feedback time that reduces the chances of mishaps. As seen above, there are many ways to gesture control drones but the method that stands out for operating in all conditions in the IMU based control as it is very sensitive to motion and it does not depend on light i.e. IMU based control can also be used in the dark but other methods rely on some form of ambient light for image processing to happen.