Advanced

User related assessment of Continuous Support & Curve Speed Control (CRF)

Varhelyi, Andras LU ; Kaufmann, Clemens LU and Persson, Anna LU (2013) SP7 working report.
Abstract
The aim of the user related assessment is to evaluate the effects on driver behaviour, reactions to and acceptance of the driver assistance system: Continuous Support (CS) function developed by CRF within the framework of SECONDS subproject. The CS function continuously evaluates the status of the host vehicle as well as the surrounding traffic based on the information of the on-board sensors. When the CS function detects a hazard, the function issues a warning to the driver. The warning status increases continuously depending on the degree of the hazard (at higher degree of hazard also sounds and active feedbacks are activated in the safety belt). The CS function provides the following warning to the driver:

- In a situation with... (More)
The aim of the user related assessment is to evaluate the effects on driver behaviour, reactions to and acceptance of the driver assistance system: Continuous Support (CS) function developed by CRF within the framework of SECONDS subproject. The CS function continuously evaluates the status of the host vehicle as well as the surrounding traffic based on the information of the on-board sensors. When the CS function detects a hazard, the function issues a warning to the driver. The warning status increases continuously depending on the degree of the hazard (at higher degree of hazard also sounds and active feedbacks are activated in the safety belt). The CS function provides the following warning to the driver:

- In a situation with the actual speed above the speed limit, the display shows the speed limit icon.

- In a situation with too high speed when approaching a curve, as pre-warning, the visual display shows a yellow curve; as imminent warning, the display shows a red curve icon, alarm sound is activated and the safety belt is tensioned.

- In a situation with risk for a forward collision, as pre-warning, the display shows a yellow obstacle icon; as imminent warning, the display shows a red obstacle icon, alarm sound is activated and the safety belt is tensioned.

- In a situation with a vehicle in the blind spot, as pre-warning, the display shows a yellow blind-spot obstacle icon; as imminent warning, the display shows a red blind-spot obstacle icon.

Twenty four persons (13 males and 11 females) - employees at CRF (persons not directly involved in the interactIVe project) took part in the test drives. The test drivers drove twice along the test route of an approximate length of 53 km, containing rural road and motorway sections. The order of driving was balanced in such a way that every other subject drove first with the system switched off and then with the system switched on. For the following subject the order of driving was reversed. The test vehicle was equipped with logging facilities and a number of variables were logged. The test drivers were observed by means of an in-car observation method (Wiener Fahrprobe) by two observers, riding along in the car with the driver, where one of the observers (called the coding observer) studies standardised variables such as speed behaviour, yielding behaviour, lane changes and interaction with other road users. The other observer carries out ‘‘free observations’’ such as conflicts, communication and special events that are hard to predict, let alone to standardise. Questionnaires were used to assess the drivers’ comprehension of and experiences with the system, their subjective workload, the usefulness and satisfaction of the system, perceived benefits of the system, as well as willingness to have and pay for the system.

The findings revealed that the test drivers did not alter their general speed behaviour when driving with the system compared to driving without the system, see Table S.1.

It has to be mentioned, that the speed warnings frequently displayed erroneous speed limits at some parts of the route, which also was commented by the test drivers. The legal speed limit was equal to the speed limit indicated by the system of approximately 60% of the route length. The speed limit reported by the system was over the legal speed limit along 28% of the route, and below the legal speed limit along 10% of the route.

There was no change in the number or in the length of speed warnings when driving without and with the system.

The profiles of speeds and accelerations before and after a forward collision warning was triggered illustrate, that when the warning was issued, the driver had already started to decrease the speed. This indicates that the warning was issued too late to have any effect on the driver. Consequently, driver reaction time to the warning could not be analysed.

Due to curve speed warnings, the test drivers passed the roundabout with a statistically significantly (p<=0.05) lower speed when driving with the system.

There was some tendency for increased number and length of forward collision warnings when driving with the system, but the difference was statistically not significant (p=0.5).

There was no difference either in the number or in the length of the side collision warnings neither form left or right when driving without and with the system.

Driving too fast according to the situation and/or the speed limits was observed statistically significantly less often during the rides with the system active. Driving too far to the right side and dangerous lane changes were observed statistically significantly less often on the rides with the system active. The test persons chose a wrong lane when driving through an intersection or roundabout less frequently, when driving with the system active.

On the negative side, it can be noted that slightly more late adaptations of speed before intersections and obstacles were observed while driving with the system. Statistically significantly more errors regarding dangerous distance to the side were observed with the system active. Only during driving with the system active it was observed that the test persons turned with too high speed.

No major differences were found regarding speed choice when driving with or without the system. The test persons drove over the speed limit (on rural roads and on highways) on both rides. Also, the test persons drove too fast through curves and approached a roundabout or drove through it too fast, as well as they accelerated before leaving the roundabout to the same extent on both rides. Bad adaptation of speed before intersections and obstacles, driving too slowly, incorrect lane choice before intersections and sticking to own priority was observed in equal numbers on both rides, with and without the system.

No statistically significant differences between the two rides could be shown regarding dangerous distance keeping to the vehicle in front, illegal or aborted overtaking manoeuvres, correct indicating behaviour, driving too far to the left or drifting or crossing the solid line, late or hesitant lane change before an intersection, crossing a stop line at intersections or roundabouts, driving against yellow at a traffic light, yield behaviour and ignoring pedestrians/cyclists.

Regarding interaction behaviour with other road users, hardly any differences could be observed. Situations on both rides were noted where the test persons either made errors in the interaction processes or showed respectful behaviour towards other road users. On both rides, situations were observed in which the test persons did not choose the correct speed, drove without foresight or too close to other road users, showed unclear behaviour to other road users or did not behave correctly in overtaking manoeuvres. The test persons also showed respectful behaviour towards other road users on both rides by giving way in different situations or adapted their speed and lateral position well. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Book/Report
publication status
published
subject
volume
SP7 working report
pages
95 pages
publisher
InteractIVe Consortium
language
English
LU publication?
yes
id
f06f5748-1d1e-4663-9cf9-62870eda268f (old id 4467712)
date added to LUP
2014-06-23 14:08:47
date last changed
2016-04-16 08:22:21
@techreport{f06f5748-1d1e-4663-9cf9-62870eda268f,
  abstract     = {The aim of the user related assessment is to evaluate the effects on driver behaviour, reactions to and acceptance of the driver assistance system: Continuous Support (CS) function developed by CRF within the framework of SECONDS subproject. The CS function continuously evaluates the status of the host vehicle as well as the surrounding traffic based on the information of the on-board sensors. When the CS function detects a hazard, the function issues a warning to the driver. The warning status increases continuously depending on the degree of the hazard (at higher degree of hazard also sounds and active feedbacks are activated in the safety belt). The CS function provides the following warning to the driver:<br/><br>
-	In a situation with the actual speed above the speed limit, the display shows the speed limit icon.<br/><br>
-	In a situation with too high speed when approaching a curve, as pre-warning, the visual display shows a yellow curve; as imminent warning, the display shows a red curve icon, alarm sound is activated and the safety belt is tensioned. <br/><br>
-	In a situation with risk for a forward collision, as pre-warning, the display shows a yellow obstacle icon; as imminent warning, the display shows a red obstacle icon, alarm sound is activated and the safety belt is tensioned. <br/><br>
-	In a situation with a vehicle in the blind spot, as pre-warning, the display shows a yellow blind-spot obstacle icon; as imminent warning, the display shows a red blind-spot obstacle icon.<br/><br>
Twenty four persons (13 males and 11 females) - employees at CRF (persons not directly involved in the interactIVe project) took part in the test drives. The test drivers drove twice along the test route of an approximate length of 53 km, containing rural road and motorway sections. The order of driving was balanced in such a way that every other subject drove first with the system switched off and then with the system switched on. For the following subject the order of driving was reversed. The test vehicle was equipped with logging facilities and a number of variables were logged. The test drivers were observed by means of an in-car observation method (Wiener Fahrprobe) by two observers, riding along in the car with the driver, where one of the observers (called the coding observer) studies standardised variables such as speed behaviour, yielding behaviour, lane changes and interaction with other road users. The other observer carries out ‘‘free observations’’ such as conflicts, communication and special events that are hard to predict, let alone to standardise. Questionnaires were used to assess the drivers’ comprehension of and experiences with the system, their subjective workload, the usefulness and satisfaction of the system, perceived benefits of the system, as well as willingness to have and pay for the system.<br/><br>
The findings revealed that the test drivers did not alter their general speed behaviour when driving with the system compared to driving without the system, see Table S.1.<br/><br>
It has to be mentioned, that the speed warnings frequently displayed erroneous speed limits at some parts of the route, which also was commented by the test drivers. The legal speed limit was equal to the speed limit indicated by the system of approximately 60% of the route length. The speed limit reported by the system was over the legal speed limit along 28% of the route, and below the legal speed limit along 10% of the route. <br/><br>
There was no change in the number or in the length of speed warnings when driving without and with the system.<br/><br>
The profiles of speeds and accelerations before and after a forward collision warning was triggered illustrate, that when the warning was issued, the driver had already started to decrease the speed. This indicates that the warning was issued too late to have any effect on the driver. Consequently, driver reaction time to the warning could not be analysed.<br/><br>
Due to curve speed warnings, the test drivers passed the roundabout with a statistically significantly (p&lt;=0.05) lower speed when driving with the system.<br/><br>
There was some tendency for increased number and length of forward collision warnings when driving with the system, but the difference was statistically not significant (p=0.5).<br/><br>
There was no difference either in the number or in the length of the side collision warnings neither form left or right when driving without and with the system.<br/><br>
Driving too fast according to the situation and/or the speed limits was observed statistically significantly less often during the rides with the system active. Driving too far to the right side and dangerous lane changes were observed statistically significantly less often on the rides with the system active. The test persons chose a wrong lane when driving through an intersection or roundabout less frequently, when driving with the system active. <br/><br>
On the negative side, it can be noted that slightly more late adaptations of speed before intersections and obstacles were observed while driving with the system. Statistically significantly more errors regarding dangerous distance to the side were observed with the system active. Only during driving with the system active it was observed that the test persons turned with too high speed. <br/><br>
No major differences were found regarding speed choice when driving with or without the system. The test persons drove over the speed limit (on rural roads and on highways) on both rides. Also, the test persons drove too fast through curves and approached a roundabout or drove through it too fast, as well as they accelerated before leaving the roundabout to the same extent on both rides. Bad adaptation of speed before intersections and obstacles, driving too slowly, incorrect lane choice before intersections and sticking to own priority was observed in equal numbers on both rides, with and without the system.<br/><br>
No statistically significant differences between the two rides could be shown regarding dangerous distance keeping to the vehicle in front, illegal or aborted overtaking manoeuvres, correct indicating behaviour, driving too far to the left or drifting or crossing the solid line, late or hesitant lane change before an intersection, crossing a stop line at intersections or roundabouts, driving against yellow at a traffic light, yield behaviour and ignoring pedestrians/cyclists. <br/><br>
Regarding interaction behaviour with other road users, hardly any differences could be observed. Situations on both rides were noted where the test persons either made errors in the interaction processes or showed respectful behaviour towards other road users. On both rides, situations were observed in which the test persons did not choose the correct speed, drove without foresight or too close to other road users, showed unclear behaviour to other road users or did not behave correctly in overtaking manoeuvres. The test persons also showed respectful behaviour towards other road users on both rides by giving way in different situations or adapted their speed and lateral position well.},
  author       = {Varhelyi, Andras and Kaufmann, Clemens and Persson, Anna},
  institution  = {InteractIVe Consortium},
  language     = {eng},
  pages        = {95},
  title        = {User related assessment of Continuous Support & Curve Speed Control (CRF)},
  volume       = {SP7 working report},
  year         = {2013},
}