Devices to help drivers control their speed - Intelligent Speed Assistance* (ISA) & Speed Assistance Systems (SAS)


Michael Paine, Vehicle Design & Research Pty Ltd  email: mpainesyd!geemail.com (replace ! with @)

Contents

Alerts & timeline for SAS


Developments in Safer Motor Vehicles Conference - 1998

Post-publication notes and updates
Pioneering Products
News about speed control devices
Disclaimer
Letters



---

* ISA started out as Intelligent Speed Adaptation but ETSC, amongst others, prefers the term Assist (or Assistance) as it better describes the direction the technology has taken.

Alerts

Includes (italics) publications authored & co-authored by Michael Paine and also some of his presentations.

Quick Tips:
Speed limit alerts: Comprehensive speed alerts for all of Australia and New Zealand are available on the iPhone as part of the Metroview Navigator app . Also available from Google Play. Speed limits are part of the offline navigation maps that have a good route guidance, if needed. The Metroview settings allow the user to select the levels of warning. I find that an audible warning at 3km/h over the limit allows me to easily drive within the speed limit. With this enabled you can turn off speed camera alerts. Importantly Metroview works with Carplay and it can run in the background so you can receive speed alerts while running another navigator app in Carplay. Seems a good $9.99 investment to me! For NSW drivers there is also the Transport for NSW - Speed Adviser - free smartphone app.

Manual speed limiter: Many modern cars have a manual speed limiter. The driver selects an appropriate maximum speed and it is then easy to stay within the speed limit. For example, Volvos have a very easy way to activate and deactivate the system: simply press the left arrow to place in stand-by mode (not actually described in the diagram below) and press the centre button (or resume button) to activate the speed limiter at the current speed. Use the up and down arrows to increase or decrease the chosen speed in 5km/h increments (or press an hold for 1km/h increments). Press the centre button (or press accelerator pedal to floor) to deactivate the speed limiter.

2024 The Conversation: Lower speeds on local streets cut deaths and injuries by a quarter in Wales. Over 100 experts want Australia to do the same + Open Letter
2023 India Today:  You can now use Google Maps to avoid speed challan, here is how - Google Maps is enhancing road safety and preventing unintentional speeding with a new real-time speed limit feature
2023 27th ESV ADAS in your pocket – a review of the features, functions and future of smartphone-based advanced driver assistance systems - potential for highly cost-effective crash reductions through smartphone apps
2022 Euro NCAP Vision 2030 - see page 9 for planned updates to Speed Assist Systems
2022 ETSC: Opinion: will Intelligent Speed Assistance (ISA) live up to its promise? [I agree with Graziella that the EU regulations have missed a great opportunity to implement ISA that is accurate and does not annoy drivers]
2021 ETSC: Intelligent Speed Assistance specifications officially published (by EU)
2021 CASR - new reports: The contribution of various levels of speeding to fatal and serious road trauma - "The results suggest that low level speeding is not benign and should remain the focus of various road safety interventions, along with all levels of speeding"  + Collection and analysis of EDR data from crash-involved vehicles: 2020-21 summary report
2021 European Transport Safety Council: iSAFER - Intelligent Speed Assistance for European Roads
2021 CASR: Intelligent Speed Adaptation (ISA): benefit analysis using EDR data from real world crashes - lower predicted savings than previous estimates but still very substantial
2020 AAP: Impact speed and the risk of serious injury in vehicle crashes - based on the study described in the ESV paper
2019 ESV: Speeding in crashes in the United States of America: A pilot study using event data recorder information from NASS-CDS (co-author with CASR)
2019: ETSC: European Parliament backs new vehicle safety standards - including overridable Intelligent Speed Assistance.
2017 Global NCAP: Report of the Seminar on Advancing Speed Assistance Systems (updated URL) + Michael Paine's slide show.
2017 ETSC: Global NCAP looks to boost take-up of speed assistance systems worldwide
2013: 23rd ESV conference: DEVELOPMENT OF AN ASSESSMENT PROTOCOL FOR AFTER-MARKET SPEED LIMIT ADVISORY DEVICES -& + NEW NCAP TEST AND ASSESSMENT PROTOCOLS FOR SPEED ASSISTANCE SYSTEMS, A FIRST IN MANY WAYS
2012:  Old Ideas plus New Technology Equals Crash Savings - Global Road Safety Partnership conference, Bangkok
2012: Pay-How-You-Drive car insurance will reward drivers  who obey speed limits
2012:: SpeedAlert Live was available for free for iPhone and Android.(Update 2015: vanished from App Store)
2012:: ISA in the USA: 5 Dec 12 IIHS Status Report: Intelligent speed adaptation devices can reduce speeding.
2009: First International Conference on Intelligent Speed Adaptation: Sydney 10 Nov 2009. 
2009: Low Range Speeding and the Potential Benefits of Intelligent Speed Assistance  (600K PDF preprint of conference paper) - findings confirmed in
Casualty crash reductions from reducing various levels of speeding (2011 CASR report)
2009: Speed limiting trials in Australia - 21st ESV
2007: What's hot in road safety? Vehicles that obey the speed limits - ACRS Seminar,
2007: In-vehicle Intelligent Speed Advisory Systems - 20th ESV
2004: Presentation on speed control at the WHO Road Safety Day in Sydney.
2003: Presentation to Staysafe on Speed Control
1998: Why consider speed control devices for vehicles
1996: Speed Control Devices for Cars - research report for NSW government

Pioneering SAS products

2007 Review of the world's first passive ISA navigator device with a national speed limit database - the Navig8r M35 with SpeedAlert™ 99% of the time I do NOT need a navigator map. 100% of the time I DO need speed limit information.This little unit displays the speed limit and beeps if you creep over the speed limit by more than a selected amount (e.g. 2km/h). It has a SpeedAlert mode where only speed limit information is displayed and no map - reducing distraction. The built-in speed limit database covers most populated areas of Australia. I have successfully used SpeedAlert in Sydney, Perth, Adelaide, Melbourne, Canberra, Wollongong, Newcastle and Brisbane.

2008 Beware of some brands of sat-nav that have a speed limit sign displayed on the packaging. The speed limit database is very limited for some devices (eg around schools and speed cameras) or, worst still, has to be set manually by the driver. To my knowledge only the SpeedAlert system has an extensive nationwide database and time-based school zones.
Update June 2009: The latest Navig8r - the G35 has the option for an analogue speedo display - a dynamic speedo that clearly shows the current speed limit. I also like the scale of the speedo, with a maximum speed of 120km/h. Video of demonstration (10Mb WMV)

Post-publication notes, news and links

I first became aware of the remarkable road safety benefits of reducing traffic speeds while researching a project for the NSW Roads and Traffic Authority in 1996: Speed Control Devices for Cars. Subsequent presentations have included:

• WHY CONSIDER SPEED CONTROL DEVICES FOR VEHICLES? - presentation at Developments in Safer Motor Vehicles, March 1998
• Presentation to Staysafe on Speed Control - 13 Oct 03 (800K pdf)
• Presentation on speed control at the World Health Organisation  Road Safety Day in Sydney.
• In-vehicle Intelligent Speed Advisory Systems - 20th International Conference on the Enhanced Safety of Vehicles, 2007 (or Pre-print copy 1Mb PDF)
• What's hot in road safety - vehicles that obey the speed limits (1Mb PDF) - presentation based on ESV paper
• COMMENTS ON DRAFT EURO NCAP PROTOCOL FOR SPEED-LIMITATION DEVICES - March 2008. Contains draft requirements for ISA systems, based on ECE Regulation 89.
• Devices to assist drivers to comply with speed limits - January 09
• Speed limiting trials in Australia - paper for 2009 ESV conference in Stuttgart.
• Nov 09:  Low Range Speeding and the Potential Benefits of Intelligent Speed Assistance  (600K PDF preprint of conference paper)

Two key points that motorists should be aware of are:

1. Half of all fatal accidents occur at an impact speed of less than 55km/h. "Low speed" collisions can be deadly. This is due to Newton's physics and the frailty of the human body.

2. Every 5km/h above the speed limit doubles the risk of being involved in a casualty accident. In a 60 zone, travelling at 65km/h doubles the risk and travelling at 70km/h quadruples the risk.

For several years I have been calling for measures that assist motorists comply with speed limits (and, incidentally, avoid speeding tickets). GPS navigation technology is bringing that closer to reality.

One of the key findings is that, for a given set of road conditions, there is a optimum maximum speed, above which the risk of fatal accidents can be expected to rise more or less exponentially. This risk is largely due to restricted sight distances and the corresponding time available to recognise that a hazard is present and decide to take avoidance action - the risk is not greatly affected by driver ability or vehicle braking performance.
Separate lines of road safety research have produced the similar conclusions:
• Swedish research (Nilsson 1993) into speed limit changes throughout the developed world suggests a fourth power relationship between mean traffic speed and the proportion of fatal crashes. A 3% reduction in mean traffic speeds can produce a 12% reduction in fatal crashes.
• Recent analysis, by the Insurance Institute for Highway Safety , of speed limit changes within some states in the USA shows a 15% increase in fatalities per vehicle kilometre travelled for states that have increased speed limits. FAQ on speed.
• The NHMRC Road Accident Research Unit has prepared a report Travelling Speed and the Risk of Crash Involvement which found that in a 60km/h speed limit area, the risk of involvement in a casualty crash doubles with each 5km/h increase in travelling speed above 60km/h
The clear message is that it is very important that speed limits are set, and enforced, according to the actual conditions so that they do not exceed the optimum level for those conditions, otherwise a disproportional increase in fatalities can be expected. It is not advisable to set speed limits according to the speeds perceived as "safe" by motorists (for example, 85th percentile traffic speeds) - given suitable input, "expert" computer systems are a much better way to decide on appropriate speed limits - see 

(link to DOC file)

Links to ISA projects around the world

Australasia

• West Australia
• Victoria
• New South Wales
• New Zealand.
  

Europe

• Sweden
• Denmark
• Belgium
• Leeds - UK
• London - UK
• Spain
• Netherlands
• Finland
• Austria
• Norway
• France
• Hungary
• European Transport Safety Council
• European Road Safety Observatory

Canada

• 28 May 1998: Prof Murray Mackay from the University of Birmingham has estimated that the "delta V" for the crash which killed Diana, Princess of Wales, was about 80km/h - not double this speed as originally speculated by the media. Note this was an offset crash so the intrusion was greater than if the crash forces had been distributed across the full width of the vehicle (or absorbed by a safety barrier!).
• In January 1999 the New South Wales Government introduced 40km/h speed limits for motorists when passing buses picking up or setting down school children. See ' Flashing Warning Lights for School Buses' for background on this issue.
• In December 1999 the NSW Government issued Road Safety 2010 which has speed management as a key strategy.
• March 2000: There has be a promising development on this issue in the UK: The executive report for 'phase 2' of a speed limiter project is now available at http://www.its.leeds.ac.uk/projects/evsc/index.html
• Dr (now Prof) Carsten was in the media a few months ago over the issue of GPS data being used to determine local speed limits (the media picked up the item as satellites spying on speeding cars - which is nonsense!). Phase 2 involved on road trials of adaptive speed limiters, driver simulator work and computer simulations of traffic flow effects. It all looks very promising. For example they predict a 20% reduction in road crashes and an 8% reduction in fuel consumption with 'mandatory' speed limiters (ie vehicle unable to exceed the posted speed limit).
• April 2000: MUARC is also conducting a trial of speed limiters in Melbourne, using GPS technology. MUARC has summaries of several related reports: The Speed Review, Speed behaviour..., Vision Zero...,
• August 2000: IIHS(PDF) and NHTSA HTML  (or PDF) issue reports (in May 2000 and Oct 1999) showing the link between vehicle travel speeds and pedestrian injuries.
• Feb 01: French invention: A.C.S. - speed Sensitive Accelerator Pedal.
• 1 Mar 01: Intelligent Speed Management - German initiative (updated URL).
• 3 Oct 01: Swedish ISA project - 5000 vehicles in trials across Sweden.
• 13 Mar 02 : Paper on the TAC Safer Car project  now available from Road Safety 2001 conference.
• 26 Nov 02 SMH: Assume the position -- new technology pinpoints radar, cameras and more. First step towards Intelligent Speed Adaption. Blackspot.com.au.
• 9 Aug 03 Staysafe: Speed and motor vehicles - inquiry by NSW parliament road safety committee - submissions sought.
• 21 Oct 03 : Presentation to Staysafe on Speed Control - 13 Oct 03 (800K pdf)

---


Developments in Safer Motor Vehicles
Conference organised by SAE Australasia and NSW Parliament Staysafe Committee

16-17 March 1998 Parliament House, New South Wales

WHY CONSIDER SPEED CONTROL DEVICES FOR VEHICLES?

In the recently published NSW road accident statistics for 1996 it is stated that 38% of all fatal accidents were speed related. In other words, 38% of all fatal crashes could have been avoided or made less severe if the vehicle had been travelling at a more appropriate speed for the circumstances.

When we think of speed-related accidents we usually think of high-speed crashes. In fact most fatal accidents occur at impact speeds well under 70km/h. 

SPEED AND CRASHES

Probability of fatality or serious injury

Figure 1 is based on casualties to seat belt wearing drivers in car to car crashes in the USA between 1982 and 1991 (Evans, 1993). The probability of serious injury or death is plotted for a range of impact speeds (more correctly delta V - the change in velocity). Notice that the fatal curve rises steeply beyond 60km/h. .

Table 1 Probability of belted driver being killed or
severely injured in two car collisions

Delta V (km/h)	Prob. of fatality	Prob. of sev. inj. or fatality
40	0.02	0.08
50	0.05	0.16
60	0.11	0.27
70	0.21	0.40
80	0.31	0.54
90	0.41	0.65
100	0.50	0.75



Figures 2 shows the frequency distribution of impact speeds for the same data as the previous figure. Notice that half of all fatalities to seat-belted drivers occurred at less than 45km/h. Although the risk of a fatality is low at these speeds the total number of reported crashes (including non-injury crashes) in this impact speed range is much greater than at the higher impact speeds. This suggests that measures which address impact speeds in the range 30 to 50km/h are just as important as those which address higher speeds. Remember that these data are for seat belted drivers.

Crash Tests

The Australian New Car Assessment Program started offset crash testing in 1993. In 1995 the impact speed for the offset test was raised from 60km/h to 64km/h in line with international practice. This resulted in a noticeable difference in crash severity. A speed increase of 6% resulted in a 13% increase in impact energy.

While advanced features such as airbags can have a significant influence on the many crashes that occur at or below the current NCAP test speed it is likely that other factors, such as structural failure will influence the outcome of crashes at higher speeds.

These pictures show a vehicle with a longstanding safety reputation at the peak of the 64km/h offset crash test.

. 

Offset crash tests of Volvo 850 at 64km/h into a deformable barrier (courtesy Insurance Institute for Highway Safety)

These and other crash tests suggest that, at impact speeds of around 70km/h, a vehicle reaches physical limits to its ability to protect front seat occupants.
 


Screen shots from the Video "The Physics of Car Crashes" by the Roads and Traffic Authority of NSW (Marketing Dept phone 02 92186315, fax 02 92183510). These are full frontal crashes into a rigid barrier.

An aim, then, of speed management measures is to reduce impact speeds - preferably to zero (ie no impact).

Stopping distances

Stopping distances are not often appreciated by motorists. When travelling at 60km/h an alert motorist will typically travel 25m between the time that an object becomes visible (but not necessarily seen) and the time when their foot hits the brake pedal. We tend to only think of stopping distance after our foot hits the pedal!

For a given set of road conditions the impact speed depends on the initial travel speed (Paine & Fisher 1996). Using typical driving conditions it can be shown that the probability of a driver fatality is highly sensitive to the initial travel speed of the vehicle. The table shows the result of this analysis where the distance to the hazard is 50m (alert reaction time of 1.5s, heavy braking). Again the probabilities are for seat-belted drivers. .

Table 2 Relationship between travel speed and outcome for one scenario.

Travel Speed km/h	Impact Speed km/h	Prob. of fatality	Prob. of severe inj or fatality
50	0 (none)	0	0
60	19	0	0.01
70	47	0.03	0.13
80	65	0.16	0.34
90	80	0.32	0.54
100	95	0.46	0.70

The probability of fatality of serious injury is highly sensitive to initial travel speed - a few km/h can make a big difference in the outcome of a crash.



This analysis supports Swedish research (Nilsson 1993) which suggests a fourth power relationship between mean traffic speed and the proportion of fatal crashes. A 3% reduction in mean traffic speeds can produce a 12% reduction in fatal crashes. On this basis measures which reduced mean traffic speeds by 2km/h in urban areas and 3km/h in rural areas would have saved 71 fatal crashes, 342 serious crashes and 1191 other injury crashes in NSW during 1994.

A similar relationship between initial travel speed and risk of fatality exists with cars hitting pedestrians, except that, in effect,  the probability of a fatality reaches 100% at an impact speed of  60km/h (see "Vehicle Travel Speeds and the Incidence of Fatal Pedestrian Crashes" by Anderson, McLean, Farmer, Lee and Brooks, J. Acc Anal. and Prev, Sept 1997)

COMPLIANCE WITH SPEED LIMITS

Compliance with speed limits

This section discusses possible voluntary measures.

It is speculated that "unintentional" speeding makes up a large proportion of the speeding problem and that drivers could do with more assistance in keeping to the speed limit. Many cars now have speed alarms or cruise control fitted but there is currently no automatic way to inform these devices of the posted speed limit. ITS technologies give us the opportunity to provide roadside devices to transmit speed limit and other information to passing vehicles. Trials of these "automatic speed limiters" are underway in the Swedish city of Lund (Hyden 1993) and in the Nethelands (TNO).

If the road infrastructure was in place then the voluntary fitting of speed alarms and automatic cruise control could become quite popular. The price would be similar to the now-banned radar detectors and they would also help drivers avoid fines and loss of licence!

Top speed limiting

Top speed limiting is a controversial issue. Again, discussion will be confined to possible voluntary measures.

There has been a recent significant change in vehicle technology which makes top-speed limiting very cheap.

Many cars these days have electronic engine management systems. Most of these already have a top speed setting but typical settings are in excess of 200km/h. It should be relatively simple to provide a more realistic setting - it is known that some manufacturers produce chips with different speed settings for different international markets.

My point is that there is a potential market for optional speed-limited vehicles - with the appropriate chip fitted on the production line.

A speed-limited vehicle would have the same performance as an unlimited vehicle up to the preset speed. However, the engine power would drop off above that speed. A consequence is that the driver would no longer take the risk of trying to overtake at grossly excessive speeds (Plowden & Hillman, 1984 - Plowden also discusses alternatives to an instant drop off in power. For example TNO in the Netherlands is currently evaluating an actuator which increases the accelerator pedal stiffnes when a set speed is exceeded). Speed limiting would be a strong deterrent to theft and joyriding. And there is the issue of teenagers borrowing the family car! In the latter case this could be an additional application for smart keys, which adjust the seating position according to the programmed key which opened the door. The speed limiter could be set to the Australian provisional driver speed limit of 80km/h.

Optional speed limiting could attract significant insurance discounts.

Voluntary speed-limiting need not carry a stigma. After all, Formula 1 racing drivers have resorted to speed limiters to avoid the severe penalties of speeding in the pit lane.



Speed-limited in pit-lane (courtesy www.f1-photos.com)

.

.

.

REFERENCES

Evans L (1993) 'Driver injury and fatality risk in two-car crashes versus mass ratio inferred using Newtonian Mechanics', Proceedings of 37th Conference of Association for the Advancement of Automotive Medicine, San Antonio, Texas.

Hyden C (1993) 'The Speed Limiter- a Simple and Potentially Efficient Safety Device', Proceedings of 26th International Symposium on Automotive Technology and Automation, Aachen, Germany.

Nilsson G (1993) 'Relationship between speed and safety: calculation method', The Speed Review: Appendix of Speed Workshop Papers, Federal Office of Road Safety, Report CR127A, Department of Transport and Communications, Canberra.

Paine M & Fisher A (1996) ' Flashing Warning Lights for School Buses', Proceedings of 15th International Conference of the Enhanced Safety of Vehicles, Melbourne.

Plowden S & Hillman M (1984) Danger on the road: The Needless Scourge, Policy Studies Institute, London.

RTA (1998) Road Traffic Accidents in NSW-1996, Statistical statements for the years 1996 & 1994.

Links to WWW sites with offset crash test data: http://www4.tpg.com.au/users/mpaine/ncaplist.html
 

Disclaimer

Any views expressed in this report are those of the author and do not reflect the views or policy of any government agency or other organisation. It is provided on the Internet in the interests of disseminating road safety information. While the information presented is believed to be accurate no assurances can be given in this regard.

Letters



• 25 Oct 03: Following a ridiculous statement about scrapping speed cameras in my local paper, the Manly Daily, from a local member of parliament I wrote to the editor to point out the risk of exceeding the speed limit. So far (a week later) the editor has not published my letter. Here it is:

• 1 May 04

• 8 May 04

Prepared by Michael Paine March 1998.  Updates - see News.


Road safety news and links