Self-driving race car drove straight into a wall off the starting line
October 31, 2020
By Paul Homewood
Whoops!!
- An autonomous race car crashed immediately during a Roborace event on Thursday.
- The car, fielded by Acronis SIT Autonomous, made a sharp right turn and accelerated into a wall right off the starting line.
- Roborace is still ironing out the kinks of its racing series, and just started its “Season Beta.”
Although driver-assistance technology has advanced considerably, we’re still a long way away from cars that can truly drive themselves on public streets. Both Tesla’s Autopilot and Chevrolet’s Super Cruise — the two most advanced driver-assistance systems — require full human attention and can’t be used everywhere.
And a recent mishap at an autonomous racing event showcased just how challenging self-driving technology can be.
https://news.yahoo.com/self-driving-race-car-drove-222400685.html
It reminds of when I played with Scalextric as a kid!
But the idea we’ll all be driving around in these things in the foreseeable future is for the birds.
60 Comments
Comments are closed.
NOT A LOT OF PEOPLE KNOW THAT 
Technologies such as wind turbines, solar panels and batteries have a practical upper limit to how much more efficient they be made. I would suggest that sensors combined with artificial intelligence do in fact have the potential to make self driving cars a reality. I don’t think that it will be any time soon but if there is a will to do it, the technology will continue to improve. I think that people will expect it to be close to 100% safe before it will be accepted. In practical terms road safety would improve provided that your self driving car was safer than an idiot.
A good book on the subject is ‘Why we Drive’ by Matthew Crawford. He explores driving as one of the activities that allows human beings a full expression of their physical and mental skills. Why do we want to devalue the human experience by automating things that we’re already pretty good at – unless there is a clear benefit to our lives in doing so?
Because we are actually pretty bad at it, given the fatality and injury statistics.
The 1,300 deaths in the UK alone? The tens of thousands of accidents with costs in all sorts of areas? The reduction in congestion and journey times? The ability to spend journeys doing other stuff? The potential savings of not having to own an expensive asset?
1870 road deaths per year (rolling year to June 2019). That is about 36 per week.
Contrast that with around 2,000 deaths per week from Influenza & Pneumonia (5 year average for the first 42 weeks of the year – ONS, pre-Covid).
Or 10,217 deaths per week for all causes.
Driving is a small cost for a very significant benefit. That’s why its so popular.
Wind turbines are not a technology. They do not work. They do not produce 50/60Hz energy. They cannot boil a jug.
The late great George Carlin said that ” The average person/driver is fairly stupid , so half of them are worse than that “.
So , while self driving cars might be better than some of the people on the road , they are nowhere near as good as the more experienced drivers .
There was a TV program with Guy Martin where they were trying the latest technology in a self driving race car . Although the program would not say , the actual lap times were hopeless compared to a real race driver .
I believe that the self driving stuff will ignore a traffic sign if it has any graffiti on it . There have been several crashes ,[ including fatalities ] , because the road markings were not as good as normal , [ so what happens with a little snow , heavy rain or road resurfacing etc ].
And cases where the vehicle has emergency braked because a pigeon or pheasant has flown in front of it .
So ,while it is possible that autonomous cars might be better than some drivers , overall they will bring the driving standards down as most drivers will not get the experience necessary to become a good driver .
An apt metaphor for the governments green recovery strategy!
If it can’t stay on a road that wide it’s useless.
Because no human driven racing car has ever hit a wall? How many crashes are there in a typical F1 weekend? Four? Five?
Yes, in race conditions, going at high speed. Especially on the start grid, and when driving at the limit, during cornering, overtaking etc.
In the video above the car just pulled away from a standing start on an empty track…straight into the wall. Never seen that in F1 racing before. Maybe from children in pedal cars.
Still, you have to start somewhere I suppose.
Not compatible in racing situations
We have the science, we have the technology, absolutely nothing can go wrong, can go wrong, can go……..
And if and when there are ‘driverless’ vehicles on the road how much will the insurance be? Who pays for any crashes — the owner, the manufacturer, the hardware or software developers, etc.?
Well going with the ‘greenie’ theme maybe ‘the government’ should [/sarc-off]
Excellent! It has reached the standard of human racing drivers and they are only at beta, well, gamma really.
Is the driver or the motor manufacturer responsible for the crash if the car’s auto driving equipment can be blamed?
Motor insurance problems galore if/when such a crazy idea for unsafe and avoidable technology is adopted, if ever.
It’s quite bizarre to believe that humans who can be tired, drunk, emotional, distracted, on the phone & only have limited vision are somehow more safe than a machine that can be none of those and which has vastly better senses constantly covering much more volume. No doubt pilots said the same thing about autopilots yet they have saved hundreds of thousands of lives.
Phoenix,
Can you substanciate that last sentence.
You are aware that pilots always manually fly for take off and landing, except the occasional restricted vision landing.
The autopilot does the menial part of flying the aeroplane in a highly regulated and open space.
It is a long and proven technology but even that goes wrong, 737 Max 8 is a good reminder.
Would you even get in a fully autonomous aircraft with no over riding control from a pilot?
Phoenix44:
“humans who can be tired, drunk, emotional, distracted, on the phone & only have limited vision are somehow more safe”
How does any of that relate to motor racing? You think Lewis Hamilton starts form Pole position tired, drunk, emotional or distracted? Or perhaps he’s tweeting just before dropping the clutch?
The video is an autonomous racing car on an empty circuit. It pulled away directly into a wall. Not even a single mitigating factor.
Evidence fot those figures/
100% driverless cars might work, but a mix of unpredictable humans and driverless will never work……if we are confronted with a car coming towards us on the wrong side of the road, we will swerve to the left or right and hope for the best, a driverless will just stop and wait for the impact.
Why would it do that? Airplanes on auto-pilot don’t do that, never did. Existing driverless cars have a number of collision avoidance strategies and none of them include “just hit the car coming towards you”.
The fail safe mode of electric cars is to hand control back to the “driver”, the electric car has to take instant evasive action, the car on the wrong side of the road is driven by a drug crazed thief escaping from a police pursuit, does the electric car swerve left or right to get out of the way? It cannot have instinct built in to its algorithms.
It doesn’t have ‘instinct’ built into its algorithms, it has a far more accurate measurement of space and trajectory and can compute outcomes of different strategies in a way that humans cannot. It will “hand back control to the driver” at the moment, but won’t stay that way.
You keep postulating situation which you assume human drivers handle well, but offer no evidence to support that assumption.
Honestly, anybody would think autonomous vehicles somehow program themselves, ignoring what humans do when they drive. The opposition to driverless cars is bizarre with some very strange objections made. This one is quite bonkers and seems to assume that a driverless car, with vastly more ability to see what’s around it in all directions and to compute more accurately and more quickly speeds and vectors will just do the dumbest thing.
Yes Phoenix, airliners would never crash because of their automated systems doing dumb things would they…… oh no, wait.
Driverless cars are fantasy for the foreseeable future. I am still waiting for my personal flying car (promised since the 1930s). Driverless railway or tram systems (like at airports between terminals) maybe. Much more controlled environment.
More automated safety systems such as collision avoidance, super duper sophisticated cruise control etc in driver cars? Absolutely. Big role to play.
Computers are very good at making high speed computations. That’s why they are called computers. Making decisions? Not a patch on a human. That’s why military jets are designed unstable with “fly-by-wire” controlled by computers but still have pilots. Even drones still have pilots.
Remote piloting of trains and trams, like military drones? Seems entirely plausible. Remote piloting of your personal vehicle? What’s the point – you will be in it going somewhere anyway.
I was once faced with a car pulling out of a side road in front of me, a foot drop where the verge on my side had been washed away, and a car coming in the opposite direction (all on a narrow road). I still don’t know how I worked it out in a split second, but I avoided an accident by swerving on to the VERGE on the OPPOSITE side of the road! As a retired software engineer (Air Traffic Control) I would hate to have to work out an expert system that could recognise a ‘potentially fatal’ situation and then do something ‘illegal’!
Excuse me if I am missing anything but the whole point of motor racing is the test of the combination of drivers skins and care mechanics. If it comes down to just cars with no driver OR a driver with a playstation….where is the entertainment spectacle? I “thought” it was driver first car second in the entertainment stakes but all is up for grabs in the future dystopian marxist nirvana.
‘Self-driving’ seems an exagerration.
What is the point ? You might reasonably state that Lewis Hamilton has the best F1 car but his driving skill makes the vital difference. This risk averse mentality will make Neanderthals of us all.
Seriously? Half the drivers in F1 at least would win with that car. The difference between the Mercedes and te rest is far greater than the difference between Hamilton and say Verstappen (who might be better anyway).
PMFB,
Absolutely!
Phoenix,
Such as Valteri Bottas, yes he has won grand prixs with the same car, Hamilton wins championships, that is the difference.
You can’t judge who is the best driver on an F1 team, there is a hierarchy.
No2 is always held back by various means/consequences.
One unintended consequence of self driving cars will surely be the deterioration of driving skills that will follow. At present driving ability is honed constantly as we drive.
Same with the first self braking car. Bang!
“But the idea we’ll all be driving around in these things in the foreseeable future is for the birds.”
Have they given up on flying?
I don’t think there’ll be much driverless driving in sub-Saharan Africa any time soon.
I used to drive a lot on business. Up at 5:30, rolling by 5:45, meeting at 9:30, somewhere. Do the day and then drive home again. I remember my first SatNav, I was on the A43 to Kettering the road signs said. Did I want to go towards Kettering? For the first time I didn’t have to worry, just drive the car and relax.
Having a self drive car that I could get into after several hours of exhausting battles in meetings and just say “Take me home” – it would have been wonderful.
To all those who say “It will never reach the standard of human drivers” – have you never to Birmingham? Have you never seen the M6 round Manchester at 6pm? Have you never encountered elderly drivers failing to cope in rush hour traffic? Those and so many other places/situations.
Unintended consequences. Have they failed to remember that with zero carbon emissions there won’t be any rubber, no asphalt, concrete or cement to drive or land anything on? Those are all bad for the climate.
Yes, but did it win?
The car, fielded by Acronis SIT Autonomous
Or Acrimonious as they try to work out whose fault it was.
Science fiction has, for years, had “auto-driving” motorways. That is, motorways have control strips, and specially equipped cars can join them, disconnecting the driver. Once on, and under central control, cars are spaced and can run at maximum speed. But that would only be of use on loooong inter-city routes – not much use in UK!
“can run at maximum speed”
As determined by the car in the lead.
An additional point to that comment: motorway safety is largely determined by appropriate following distance, not speed. Plus looking (or not looking) when changing lanes.
Autonomous systems alerting driver and possible activating braking when following distance is changing rapidly – definitely. Also need to take account of vehicle behind speed and following distance. These systems would work.
On a country road when a deer jumps out? Unlikely. Unless every deer has a GPS collar on it.
A self-driving tank might be handy. Especially if it were amphibious.
I really would like to see a self driving car on our country roads up here in Pembrokeshire, every year we get an influx of visitors seemingly incapable of rising to the challenge.
Maybe it was the equivalent of a “half time” show?
Christian Wolmar’s book ‘Driverless cars: On a road to nowhere’ is an excellent read. What makes it good is that Wolmar is no dewy eyed ‘technology is the way to go person’. He puts forward a key element of this driverless lunacy in that it is not car companies that are behind it but techie mobs probably trying to show how clever they are. For those who think every new technology idea is great there is always the question of ‘how is your 3D TV?’ or ‘how is that curved screen TV?’.
Must have been using a green computer
October 2020 mean CET is in – anomaly 0C for 1961-1990 reference period, ranked 94th warmest (with 10 other years) out of 362.
Temperature may have been very average, but it was notably wet in some places, I’m in quite a dry area of the country, we had 195.4mm (7.7″) of rain, 223% of average.
CET Anomaly for the last 2 months needs to be under about +2.1C to avoid a yearly hottest ever, getting much less likely but still possible, but only if the long-term weather outlook completely changes.
Is your windows software running slow?
Has anyone seen the latest load of “tripe” (that’s putting it politely, and an insult to tripe) from Rowatt on the BBC website today.
Talk about uninformed wind.
Totally off topic but you’ll be interested in this
Could Scotland ever be ‘the Saudi Arabia of renewables’?
Back in 2009, only 27.2% of Scotland’s electricity came from renewable energy sources. It was 90.1% in 2019.
https://www.bbc.com/news/uk-scotland-scotland-politics-54441613
Human-piloted race cars occasionally do this too. 😉
Oxford, wet October, but not as wet as 1875.
Only 1 of the top 10 wettest days has occurred this century. Don’t panic!
https://www.bbc.co.uk/news/science-environment-54783196
Sorry….. wettest months (not days).
Picked a heck of a time to do that!
