There are an abundance of companies pushing the boundaries, driving research and development into autonomous automobiles, also known as robotic or driver-less vehicles. These well-known brands have acknowledged the potential of autonomous vehicles to improve transport efficiency and road safety and in response to conjecture have begun to develop autonomous and crash avoidance technologies that when integrated in a combined suite allow the vehicle to become truly “driver-less/autonomous”. These vehicles have the potential to completely change the relationship between commercial entities and their fleets (for example, car rental companies in Sydney) as well as redefining how we privately use our vehicles . Over the following weeks several of these companies, namely Google, Lexus and Ford, and the technologies behind the development of their autonomous vehicles will be analysed. And, as we shall see, the variety of methods used to achieve autonomy are remarkable. The focus for this week will be on how Google has revolutionised the concept of autonomous vehicles with their fleet of robotic Toyota Pruises that, as of last year, had logged more than 300,000 kilometres on American roads. There will also be a brief discussion regarding the kinds of electronic technologies that are necessary to achieve this feat. To achieve this seemingly impossible task Google uses a range of new electronic systems and sensors, at the heart of which lies a laser range finder mounted onto the roof of the car. The device generates a detailed 3D map of the environment. The car then combines the laser measurements with high-resolution maps of the world, producing different types of data models that allow it to drive itself while avoiding obstacles and respecting traffic laws. The other sensors in the suite compliment this technology by performing a range of other tasks necessary to provide sufficient information for the vehicle to function autonomously. These include front and rear radars that allow cars to deal with the traffic conditions of freeways, characterised by fast traffic and over long distances, a GPS, inertial measurement unit, and wheel encoder, that determine the vehicle’s location and keep track of its movements. There are two interesting factors in Google’s approach. Firstly, the use of extremely detailed maps of the road and terrain is something that, according to Google, is essential to determine accurately where the car is, as by using GPS-based techniques alone the location could be off by several meters, which is obviously absurdly unsafe for use in a real world environment. Secondly, their system requires a large amount of input detail regarding its surroundings. Before sending the self-driving car on a road test, Google engineers drive along the route one or more times to gather data about the environment. When it’s the autonomous vehicles turn to drive the same route it compares data acquired during previous tests with the current input to determine its exact location. This approach is useful to differentiate pedestrians from stationary objects like poles and mailboxes. Whilst still far from being commercially viable, there is little doubt that Google has developed and produced a software/hardware suite capable of not only navigating through but negotiating with relative ease complex situations that arise in city traffic, on mountainous roads and busy highways and all with only the occasional human intervention. The crash avoidance technology and the potential for semi-autonomous to completely robotic vehicles will play an important role in the future makeup of large commercial fleets, such as car rental companies, where the majority of accidents occur. In conclusion, although this may not be the only technological/electronic method of producing an autonomous car, but it certainly is a successful method, practically oozing ingenuity in addition to its proven track record.