Telematics monitoring your car

[Ancient]

Anyone with any experience of this?

My company car is supplied by a fleet management company who are experimenting with fitting telematics monitors into their cars. Mine is one of the trial cars and coincidentally the system was fitted a couple of days before the ADUK day in Brecons.

For that day my overall score in the telematics system was 20% (i.e. in the "red" area, above 75% is required to be a safe "green"). On the morning drive with Gareth, the score was 40.1 and in the afternoon with John, it was 6% . The system measures forces applied to the car to left, right, acceleration and braking. Now I did get some suggestions that I could be smoother coming off the brakes and I might try using fixed grip more (too much driving a Vitara - short footwell and long legs!), but this worried me. The details for the day's journeys showed no 'red' incidents, but it appears not to like Brecon corners nor apparently, roundabouts.

Getting in touch with the system's suppliers, I asked why these journeys were flagged 'red' and was told that there doesn't need to be any 'bad' events, but the system scores on the number of sharp turns per mile. Apparently other people suffer because they live in areas with many mini-roundabouts (sharp turn at 10 - 15 mph)!

Any experience with these out there? Comments on their contribution to road safety?

[jont]

The key would be to understand how they have arrived at their reference values for "safe" driving. It appears they may have just picked some arbitrary numbers with no proper correlation between those and "safe" driving.

ISTR when Martin and Tony were involved with the young drivers project using telematics they went out to calibrate it with some IAM-style driving to get meaningful thresholds before putting the kit in youngsters cars.

[jcochrane]

I found this a very interesting post. I have wondered about their use in cars to monitor how safe a persons driving is. I don't know much about them but the little I had heard you have already mentioned in your post. Your findings would appear to confirm the reservations I had about them. As jont points out the correct calibration is important.

Although suggestions were made about smoothness at no times was it unsafe or that un-smooth to have such an apparent effect on the results.

[shinny]

I don't really understand how smoothness and gentle cornering corresponds to safe driving in the way these units seem to imply. A slow driver is not necessarily safe, just as a fast driver is not necessarily unsafe. It always feels like these systems are designed to target reckless (dare I say "boy racer" stereotype) drivers by assuming anyone taking a corner with a few G forces is somehow reckless. It cannot take into account visibility or observation skills, which are much more important to safety then simply reducing G forces.

[7db]

I have been pondering this sort of thing since having a meeting with my insurers about our van insurance and the issue of trackers came up.

I think that very few accidents are caused by sharp movements and loss of control, and many more by hitting the car in front after failing to leave enough space. I wonder about a parking sensor-type technology in the front bumper, measuring distance to the vehicle ahead and factoring in speed, time to that vehicle, together with whether it is increasing or decreasing (which might be a factor of space changing or speed changing).

I'd have a simple dash light system. Green light - 3s or more. Amber light - 2-3s: flashing if decreasing. Red light - 2s or less. Flashing if decreasing. (Those times up for debate).

I'd then track amount of time in the red / amber and how much of that time is flashing (ie dangerous and driver is not doing anything to mitigate).

A blunt tool, but strikes me as a good real-world measure of the ability to maintain the safety bubble longitudinally -- and also doesn't really reflect badly on speed or acceleration per se.

[jont]

I have been pondering this sort of thing since having a meeting with my insurers about our van insurance and the issue of trackers came up.

I think that very few accidents are caused by sharp movements and loss of control, and many more by hitting the car in front after failing to leave enough space. I wonder about a parking sensor-type technology in the front bumper, measuring distance to the vehicle ahead and factoring in speed, time to that vehicle, together with whether it is increasing or decreasing (which might be a factor of space changing or speed changing).

I'd have a simple dash light system. Green light - 3s or more. Amber light - 2-3s: flashing if decreasing. Red light - 2s or less. Flashing if decreasing. (Those times up for debate).

I'd then track amount of time in the red / amber and how much of that time is flashing (ie dangerous and driver is not doing anything to mitigate).

A blunt tool, but strikes me as a good real-world measure of the ability to maintain the safety bubble longitudinally -- and also doesn't really reflect badly on speed or acceleration per se.

I'm waiting for the excuses for rear end accidents - "well I didn't want to brake sharply in case the telematics gave me a red flag..."

[GJD]

I'm waiting for the excuses for rear end accidents - "well I didn't want to brake sharply in case the telematics gave me a red flag..."

That's the bit about it that concerns me, particularly with newly qualified drivers. The idea of something niggling at the back of a driver's mind that hard use of the brakes might be counted against them is not something I'm comfortable with. Of course needing to brake very hard as a late reaction to something might well indicate an error, but it's an error of anticipation and planning and once the error is made, braking hard could very likely be exactly the right thing to do.

[jont]

I'm waiting for the excuses for rear end accidents - "well I didn't want to brake sharply in case the telematics gave me a red flag..."

That's the bit about it that concerns me, particularly with newly qualified drivers. The idea of something niggling at the back of a driver's mind that hard use of the brakes might be counted against them is not something I'm comfortable with. Of course needing to brake very hard as a late reaction to something might well indicate an error, but it's an error of anticipation and planning and once the error is made, braking hard could very likely be exactly the right thing to do.

There are similar concerns with curfews (again often targeted at young drivers) - ie "I was speeding home because otherwise my insurance would stop/double at 11pm")

[trashbat]

I've written about this before, but probably not on here.

I like the idea of telematics, not least because it's about individual driving behaviours rather than lumped demographics. I think that current technology probably isn't good enough, but what emergent tech is, and I think the current downsides shouldn't be terminal.

For example, what if you developed telematics that included video? This could use pattern recognition to determine how quickly you react to the car in front brake lights coming on, how far you generally keep from the car in front, and so on.

An emergency stop is currently counted as a black mark. Indeed there's a case to be made for that being correct regardless; someone who habitually has to brake hard is a risk. However, in an isolated incident, there is a difference between (a) you failing to spot a line of stopped traffic ahead, braking hard and missing by 1cm, and (b)the car in front doing one, you reacting quickly and performing a stop that incurs the same deceleration.

Automated video analysis is feasible today, albeit perhaps in a non-productionised way. It might not be economically viable right now, and it might be eclipsed by autonomous vehicles before it ever found a market.

Regardless, I think there are interesting developments of the base telematics we have today.

[Ancient]

I've written about this before, but probably not on here.

I like the idea of telematics, not least because it's about individual driving behaviours rather than lumped demographics. I think that current technology probably isn't good enough, but what emergent tech is, and I think the current downsides shouldn't be terminal.
...
Regardless, I think there are interesting developments of the base telematics we have today.

I agree and was glad to be part of the trial system. I am worried however by the measurements it gave of my driving on the Brecons day. As I mentioned to the supplier, I live in an area where there are frequent sharp bends, narrow roads with passing places etc. Driving safely to be able to stop within the distance I can see to be clear (or twice that distance where relevant), if I see a hazard that requires me to stop, especially an oncoming vehicle on a narrow road - then I'll stop (and take the passing place if available) and quickly! That is good driving practice; gently braking to a halt and hoping the other driver stops in time is not!
Current systems appear to favour urban dwellers with ots of straight roads to drive on (and no roundabouts). Apparently the system has been "University tested" whatever that means!

[GJD]

Apparently other people suffer because they live in areas with many mini-roundabouts (sharp turn at 10 - 15 mph)!

I've heard of that too. It should be fairly easy to solve though. As I understand it, these things generally include GPS so all you need to do is either map where all the mini roundabouts are in advance and avoid alarms (or maybe use a different threshold) there, or once you're monitoring a large enough fleet of vehicles you can look for patterns: e.g. locations where lots of people are using more g but low speed and it's not leading to problems.

Not difficult, but it takes time for things like that to develop, and I'm sure it's annoying in the mean time.

[GJD]

I agree and was glad to be part of the trial system. I am worried however by the measurements it gave of my driving on the Brecons day. As I mentioned to the supplier, I live in an area where there are frequent sharp bends, narrow roads with passing places etc. Driving safely to be able to stop within the distance I can see to be clear (or twice that distance where relevant), if I see a hazard that requires me to stop, especially an oncoming vehicle on a narrow road - then I'll stop (and take the passing place if available) and quickly! That is good driving practice; gently braking to a halt and hoping the other driver stops in time is not!

That's one example of a general problem I'd imagine there'd be a lot of: that of having one's driving judged by someone else's standards and measuring tools rather than one's own.

My insurance company has never mentioned telematics to me. If they did, I think I'd politely decline. I tried to add a named driver once for a day of advanced driving coaching and they assumed that meant drifting, skids, handbrake turns and such like. I was not polite on that occasion. That level of ignorance puts them way off my list of people whose opinions I have any interest in when in comes to assessing my driving.

[Mr Cholmondeley-Warner]

Driving safely to be able to stop within the distance I can see to be clear (or twice that distance where relevant)

Perhaps you could explain where twice the distance would come into play? *










* Assuming you meant half

[7db]

Where you're driving a really big truck that people reverse out of the way of?

[Ancient]

Driving safely to be able to stop within the distance I can see to be clear (or twice that distance where relevant)

Perhaps you could explain where twice the distance would come into play? *










* Assuming you meant half

Confusing the opposites and thinking of where I can see twice the distance I need to stop.