Specifications

From all of these details it seems that a very successful method of judging proximity,

and also possibly recognising distance, is to use an infrared LED (Light Emitting

Diode) and an infrared detector. When the robot is facing a wall or object within a

certain distance (unknown at this stage and variable) infrared light will reflect from that

object and be collected in the detector. The higher the voltage or current returned from

the receiver, the closer the object must be. To overcome the possibility of natural light,

containing infrared, affecting the functionality of this sensor configuration, it will be

necessary to clock the infrared emitter at a rate of 36 – 40 kHz. The receiver will also

have to be tuned in order to only recognise this frequency of light.

2.4 Object Avoidance

Many applications are beginning to incorporate object avoidance technology into their

design. Cars are becoming more autonomous, traction control for example, and this

technology is bound to move towards object avoidance. Reversing sensors can alert the

driver when an object is close to the rear of the vehicle. It is only a matter of time

before a car will be able to recognise which lane it is and manage the steering system to

ensure it does not cross any lines. This technology is sure to save many lives from the

hazard of driver fatigue.

Object avoidance technology is currently installed into many items such as automatic

vacuum cleaners, auto lawn mowers and other items where little human instruction is

desired. Electrolux’s Trilobite automatic vacuum cleaner uses random driving while

avoiding collision in order to cover the floor space.

A project conducted by a student at Niagara Technology University used sensors and

hardware to create a object avoiding robot not dissimilar to that of this dissertation. The

concept behind this project gave the robot a human like thought process in that it would

come to an object, view the left then view right and make a decision of which way was

best to turn. If the robot had cornered itself then it would turn around and continue

driving.

David Tunnel 2004 suggested that there was no perfect solution available for

autonomous object avoiding and hence started a proposal to create an algorithm to