Site menu:

Latest News :

July 16, 2010:
The AHRS PC board has now gone through is specification and design phase. We hope to be testing this new unit soon. Tthe Extended Kalman Filter (EKF) has been written and just need to be tested on hardware.

Sept 18, 2009:
Hardware PCB Prototype is awaiting fabrication. Hoping to have board assembled and tested by mid next month. BatchPCB has a long turn around time.

Aug 13, 2009
Development of Base Station Software is coming together quickly. Working to create the most intuitive and therefore easy to use client software is our main priority but has also enabled up to create some innovative new control methods.

The Helicopter

The initial hardware system will be small and portable. With the base test bench done and working on a small 40 size scale helicopter we will have the ability to scale to a larger, more stable platform capable of carrying more weight and therefore more sensor and camera equipment.

Chassis
With the electric T-REX 450SE we will be able to conduct testing indoors as well as outdoors. Our goal is to develop a fully customizable computer system that has the ability to learn the helicopter platform response and therefore will require little external modifications when moving from one helicopter system to another.

Sensors
There are endless amounts of sensors capable of being placed on an autonomous helicopter. There are however only a few that are required to sustain autonomous flight, namely an Inertial Measurement Unit (IMU), and a Global Positioning System (GPS). The IMU consists of three accelerometers and three gyros placed to measure 3 axis movements. These Microelectromechanical (MEM) sensors are integrated and used to define the current rate of acceleration along these axes as well as the Euler Coordinates.

Flight Dynamics
Figure 1 - Image courtesy of Wikipedia

The general problem of using an IMU is it is not perfectly accurate and will eventually suffer from drift. A common solution to this problem is integrating some absolute sensors that will fix these inaccuracies. We are using a Differential GPS based system which has the ability give position coordinates down to sub centimeter accuracies. This system is not as crucial to hovering and basic flight as the IMU is, but will allow us to follow long paths and hit our destination point exactly.

All the sensors are integrated using a Kalman filter which is commonly used to estimate position in a dynamic 3D space. The IMU and integrated GPS is called our Altitude and Heading Reference System (AHRS).

Flight System
The autonomous control system is to be laid out in a simple and elegant manner and use the minimalistic approach to simplify the solution at hand. As you can see in Figure 2 there is little involved in the structure of the system. The RC receiver is the previous method of control for the initial helicopter chassis. The autonomous control system integrates up to 12 PPM signals in and is able to directly control the digital servos or split and be used for the processor. The AHRS is the core of the external system input and is used define the real-time orientation and angular rates of the helicopter, in order for the Central Processing Unit (CPU) to make accurate decisions to enable extreme stability.

System Diagram
Figure 2 - Autonomous Control System Diagram

Figure 3 is a diagram of the preliminary electronics block. We are using an Xilinx Complex Programmable Logic Device (CPLD) as the core interface between the radio receiver and the digital servos for the autonomous and manual control switchover. While in Manual mode, the CPLD will serve to copy the real time incoming data from the RC receiver and pass it to the processor in order for the on-board computer to learn the helicopter flight characteristics from brief manual flights. This will hopefully eliminate the need to set servo travel and speed restrictions based on what the pilot commands into the helicopter while it is in manual control. The use of fast logic devices eliminates lag from microcontroller circuitry and allows extreme precision on the pulse width for the digital servos while in autonomous mode.

Electronics Flow Diagram
Figure 3 - Electronics Flow Diagram

Safety
For safety and practicality there will always be two methods of control, one autonomous and the other a human pilot operator. With the pilot redundancy we have an increased chance of a crash incident. Our autonomous system would be enabled or disabled at the control of the pilot operator.

Ground Station Control
The main method of autonomous control would be through installed software on a ground base station or PDA. From within this interface one is able to define the flight type: Waypoint Mode, Fast Flight Waypoint Mode, and Stability Mode. The first two modes differ in the way a waypoint location is perceived. In waypoint mode the helicopter will navigate towards the point and will only point to the next waypoint once the helicopter is within a certain radius of the point. Using the Fast Flight mode the helicopter knows the waypoints ahead of it and will attempt to round off sharp bends between points if it is necessary to do so. Using this approach, it is easy to program a circular path to travel rather then a hexagon of some sort. In the Stability Mode the user is in real time control and is able to navigate the helicopter using simple directional commands, there is no need to worry about the details in flying a remote control helicopter, however one must maintain visual contact with the scale helicopter at all times.