futuristic autonomous car
2
Ken Gratton25 July 2018
NEWS

Why synapses beat silicon chips

Are we getting ahead of ourselves in our dependence on electronics to pilot cars?

COMMENT
Have you noticed? You won't see any computer science graduates winning grand slams or squash tournaments.

Computer scientists are working with machine 'intelligence' that can process data at many times the speed of the human brain. Yet those computer scientists themselves appear to have no conception of speed in the real-world environment – the chaos of ballistics, inertia and risk to life and limb.

Just ask someone from your help desk to defrag your hard drive and you'll quickly learn that a speedy response is 'sometime next week'.

So is it any surprise that the same IT specialists who are so risk-averse in the first place – password protection? Static charge protection? – are so deeply out of their depth when it comes to traffic.

One almost gets the impression that they would prefer everyone work from home and walk the kids to school, but if push comes to shove they will develop a level-five autonomous motoring 'brain' – a device that would be reluctant to poke the nose of the car past the nature strip.

We were all so impressed by the vision several years ago, of a Google-guided Prius ripping around an autocross course at a speed that was close to tearing tyres off the rim.

But realistically, that's nothing like a conventional traffic environment.

image credit: www.pinterest.com.au/jtr1264/

Anyone – possibly even my good self – could type up a batch file for Windows 3.1 to run a slot car around a Scalextric track at a decent clip without the car tipping over or spearing off the track.

But ask any software developer to work out a guidance system that can reconcile the random movements of pedestrians and cyclists, or react instinctively (and promptly) when an oncoming car unexpectedly veers across in front of you – and it's like shining a light on dark matter.

Computer processors are just not intelligent enough to predict anarchic traffic movements. It only takes a week or two in an AEB-equipped car to realise that every new situation on the road has to be analysed anew by the processor – without regard for learning from previous experience. There's no ability to learn, no ability to extrapolate and no ability to put such learning into practice as a countermeasure. Not yet, at least.

Those people who put so much faith in their Teslas and Uber-Volvos that they literally leave the vehicle to its own devices are likely the sort of computer-savvy dunderheads who should never be allowed behind the wheel of a real car anyway.

Share this article
Written byKen Gratton
See all articles
Our team of independent expert car reviewers and journalists
Meet the team
Stay up to dateBecome a carsales member and get the latest news, reviews and advice straight to your inbox.
Subscribe today
Disclaimer
Please see our Editorial Guidelines & Code of Ethics (including for more information about sponsored content and paid events). The information published on this website is of a general nature only and doesn’t consider your particular circumstances or needs.
Scan to download the carsales app
    DownloadAppCta
    AppStoreDownloadGooglePlayDownload
    Want more info? Here’s our app landing page App Store and the Apple logo are trademarks of Apple Inc. Google Play and the Google Play logo are trademarks of Google LLC.
    © carsales.com.au Pty Ltd 1999-2026
    In the spirit of reconciliation we acknowledge the Traditional Custodians of Country throughout Australia and their connections to land, sea and community. We pay our respect to their Elders past and present and extend that respect to all Aboriginal and Torres Strait Islander peoples today.