Project Documentations

Octocopter

This unique octocopter is basically an extension of hexacopter but this model is weight efficient and there comes the lifting capability.

For this we start with the aluminium rods the lighter material to design the frame, we used two batteries for a great power supply for the eight motors. The symmetry of the frame decides the flight.

Human Following Drone:

Creating a completely autonomous drone is a quite hard task, involving a series of problems in the field of:

  • control theory (to control the propellers)
  • robotics (to design the drone and its hardware)
  • planning (to decide which path to follow)
  • vision and perception (to understand the environment)

we are going to focus on object detection, using the recent breakthroughs of deep learning. implement a simple version of person detection and following using an object detection model Machine Learning API

Warehouse Inventory

Autonomous Indoor Navigation of Unmanned Aerial System (UAS) poses many challenges. Our main reason is that GPS has limited precision in an indoor environment. The task becomes more challenging when we use additional Navigation systems, such as a range finder which consumes much power, carrying weight and adds extra cost. For estimating the attitude of the drone indoors. We are combining data from both the optical flow and the camera for following the line. The drone can also track some features of the line and make corresponding path changes.

Autonomous Hexacopter (4i-Labs)

Our objective is to design a mechanism to land rockets. Most rockets that are currently used across the world are designed to fly only once. After the rocket takes the payload to its orbit it falls back to earth into the ocean getting completely destroyed. We are trying to design a system which lands the rocket safely back onto earth so that it can be reused. The only cost for each mission would be the fuel. This would reduce the cost of mission by a factor of 5.

Autonomous Hexacopter (4i-Labs)

A six armed UAV capable of flying autonomously just by receiving destination address. 2D Lidar is used to generate 2D surrounding map which in turn is used for routing. RGB-D images are used for obstacle avoidance. Pixhawk flight controller is used to control the UAV during the flight. Messages are sent and received between the pixhawk and ground station/On board computer over telemetry/wired connection. Generates real scale maps using location stamped images from GoPro and performs human counting as well. All the code is run onboard using Intel NUC. It is accompanied with Intel Compute Stick 2, all the Neural network processing is offloaded to it. Capable of flying even in case some motor stops working. All these functionalities find great use in the case of military surveillance, disaster safety, agriculture, poaching prevention, delivery services etc.

RC Plane

We have modified the basic design of an RC plane to include cooling mechanism and package carrying mechanism.

  • Total length (motor’s tip to elevator’s end) = 63.5 cm
  • Lateral thickness (thickness measured horizontally) = 5.2 cm
  • Vertical thickness = 5.2 cm
  • Elevator’s max length = 4 cm
  • Rudder’s length = 13.2 cm
  • Wingspan = 93 cm
  • Porthole (situated above the fuselage) = 3.7 cm x 2.6 cm

Object Avoiding Drone

This is an obstacle detecting and avoiding drone which is operated manually but the drone starts working autonomously if an obstacle is detected. The drone is mounted with an on board computer which operates the four ultrasonic sensors. The on board computer (Raspberry Pi 3) communicates with the flight controller(Pixhawk) using Mavlink protocol and takes over the control from user when it detects an obstacle within a range of 1 meter. The drone can tackle multiple objects at a single time from different directions.

Hand Gesture 2.0

We used the MPU6050 six axis IMU sensor to measure the Roll, Pitch and Yaw movements of the hand.

We added a potentiometer for throttle control and a push button for arming and disarming purposes.

NRF24L01 modules were used for transmitting and receiving the signals. One was on the hand and the other was placed on the quadcopter as a receiver.

From the previous version the ability to control the Yaw of the drone was added. This was quite a challenging task as there is no magnetometer, hence we had to rely on the data from the gyro to achieve this.

Arduino Flight Controller

A flight controller is a small circuit board of varying complexity. Its function is to direct the Rpm of each motor in response to input.

A command from the pilot for the multi-rotor to move forward is fed into the flight controller, which determines how to manipulate the motors accordingly.

Majority of flight controllers also employ sensors to supplement their calculations. These range from simple gyroscope for orientation to barometers for automatically holding altitudes. GPS can also be used for auto-pilot or fail-safe purposes.

A flight controller is pre-programmed by many ranges of firmware (software as in a computer).

Many options are available which can be tweaked using a flight controller configurator and can be set to our desired needs.