JS
J.C.J. Stapel
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
Pedestrians today are very vulnerable on urban roads. Clear communication between drivers and pedestrians is one way to reduce their plight. Non-verbal communication in particular plays an important role in road safety, and eye contact is a kind of non-verbal communication that has the potential to minimize on-road collisions. However, with the advent of automated vehicles, driver-pedestrian eye contact loses its meaning since there is no longer a driver. It is therefore useful to study and detect eye contact so that the knowledge obtained may be applied to automated vehicles of the future. To this end, the following research goals were adopted: (a) What is eye contact between a pedestrian and a driver in a car? How can eye contact be defined/operationalized using an algorithm?, (b) How accurate is the algorithm that operationalizes eye contact?, and (c) How is it possible to use two eye-trackers with inertial measurement units (IMUs) and pedestrian recognition in a Toyota Prius car to reconstruct the entire driver-pedestrian interaction through a 3-D animation? An indoor experiment, designed to resemble a driver-pedestrian interaction at a pedestrian crossing was conducted with 31 participants. Participants’ (pedestrians’) eyes were tracked using a Tobii Pro Glasses 2 eye-tracker and the researcher’s (driver’s) eyes were tracked using a Smart Eye Pro dx eye-tracker,
both of which were synchronized. Participants’ locations were also tracked using a stereo camera equipped with pedestrian detection capabilities. Pedestrians imagined that they were on a real road and performed six types of trials
where they stood on / crossed from the left / right side curb in front of the stationary vehicle while either making eye contact or not making eye contact with the driver. The order of the trials was randomized, and each trial consisted of 3 repetitions of a driver-pedestrian interaction. If the driver and pedestrian were looking at each other at the same time there was eye contact, otherwise there was no eye contact. Significant differences in the percentages of eye contact between pedestrians standing on the left (median duration of 0.42 s) and the right (median duration of 0.54 s) were found. No significant differences in the percentages of eye contact between pedestrians crossing from the left (median duration of 1.23 s) and the right (median duration of 1.39 s) were found. Eye contact instants within trials were algorithmically detected by finding the angle between the 3-D gaze direction vectors of the driver and the pedestrian, and comparing it to an ‘eye contact threshold’. Trials were classified as either involving eye contact or not involving eye contact based on their percentages of eye contact instants. The classification performance of the algorithm was quantified using two ground truths: (1) Imposed eye contact (in half of the trials, participants were instructed to make eye contact; in the other half, participants were instructed not to make eye contact), and (2) Manually annotated areas of interest (AOIs) from the Tobii Pro Glasses 2 showing pedestrian eye contact seeking. The algorithm’s performance was found to be fair/poor and eye contact could be detected with an accuracy of 15-30°. A 3-D reconstruction of the driver-pedestrian interaction was achieved (in the form of an animation) by using the locations, head orientations and gaze directions of the driver and the pedestrian. This thesis provides objective measurements of driver-pedestrian eye contact and demonstrates how eye contact may be detected and reconstructed for use in automated vehicles of the future. ...
both of which were synchronized. Participants’ locations were also tracked using a stereo camera equipped with pedestrian detection capabilities. Pedestrians imagined that they were on a real road and performed six types of trials
where they stood on / crossed from the left / right side curb in front of the stationary vehicle while either making eye contact or not making eye contact with the driver. The order of the trials was randomized, and each trial consisted of 3 repetitions of a driver-pedestrian interaction. If the driver and pedestrian were looking at each other at the same time there was eye contact, otherwise there was no eye contact. Significant differences in the percentages of eye contact between pedestrians standing on the left (median duration of 0.42 s) and the right (median duration of 0.54 s) were found. No significant differences in the percentages of eye contact between pedestrians crossing from the left (median duration of 1.23 s) and the right (median duration of 1.39 s) were found. Eye contact instants within trials were algorithmically detected by finding the angle between the 3-D gaze direction vectors of the driver and the pedestrian, and comparing it to an ‘eye contact threshold’. Trials were classified as either involving eye contact or not involving eye contact based on their percentages of eye contact instants. The classification performance of the algorithm was quantified using two ground truths: (1) Imposed eye contact (in half of the trials, participants were instructed to make eye contact; in the other half, participants were instructed not to make eye contact), and (2) Manually annotated areas of interest (AOIs) from the Tobii Pro Glasses 2 showing pedestrian eye contact seeking. The algorithm’s performance was found to be fair/poor and eye contact could be detected with an accuracy of 15-30°. A 3-D reconstruction of the driver-pedestrian interaction was achieved (in the form of an animation) by using the locations, head orientations and gaze directions of the driver and the pedestrian. This thesis provides objective measurements of driver-pedestrian eye contact and demonstrates how eye contact may be detected and reconstructed for use in automated vehicles of the future. ...
Pedestrians today are very vulnerable on urban roads. Clear communication between drivers and pedestrians is one way to reduce their plight. Non-verbal communication in particular plays an important role in road safety, and eye contact is a kind of non-verbal communication that has the potential to minimize on-road collisions. However, with the advent of automated vehicles, driver-pedestrian eye contact loses its meaning since there is no longer a driver. It is therefore useful to study and detect eye contact so that the knowledge obtained may be applied to automated vehicles of the future. To this end, the following research goals were adopted: (a) What is eye contact between a pedestrian and a driver in a car? How can eye contact be defined/operationalized using an algorithm?, (b) How accurate is the algorithm that operationalizes eye contact?, and (c) How is it possible to use two eye-trackers with inertial measurement units (IMUs) and pedestrian recognition in a Toyota Prius car to reconstruct the entire driver-pedestrian interaction through a 3-D animation? An indoor experiment, designed to resemble a driver-pedestrian interaction at a pedestrian crossing was conducted with 31 participants. Participants’ (pedestrians’) eyes were tracked using a Tobii Pro Glasses 2 eye-tracker and the researcher’s (driver’s) eyes were tracked using a Smart Eye Pro dx eye-tracker,
both of which were synchronized. Participants’ locations were also tracked using a stereo camera equipped with pedestrian detection capabilities. Pedestrians imagined that they were on a real road and performed six types of trials
where they stood on / crossed from the left / right side curb in front of the stationary vehicle while either making eye contact or not making eye contact with the driver. The order of the trials was randomized, and each trial consisted of 3 repetitions of a driver-pedestrian interaction. If the driver and pedestrian were looking at each other at the same time there was eye contact, otherwise there was no eye contact. Significant differences in the percentages of eye contact between pedestrians standing on the left (median duration of 0.42 s) and the right (median duration of 0.54 s) were found. No significant differences in the percentages of eye contact between pedestrians crossing from the left (median duration of 1.23 s) and the right (median duration of 1.39 s) were found. Eye contact instants within trials were algorithmically detected by finding the angle between the 3-D gaze direction vectors of the driver and the pedestrian, and comparing it to an ‘eye contact threshold’. Trials were classified as either involving eye contact or not involving eye contact based on their percentages of eye contact instants. The classification performance of the algorithm was quantified using two ground truths: (1) Imposed eye contact (in half of the trials, participants were instructed to make eye contact; in the other half, participants were instructed not to make eye contact), and (2) Manually annotated areas of interest (AOIs) from the Tobii Pro Glasses 2 showing pedestrian eye contact seeking. The algorithm’s performance was found to be fair/poor and eye contact could be detected with an accuracy of 15-30°. A 3-D reconstruction of the driver-pedestrian interaction was achieved (in the form of an animation) by using the locations, head orientations and gaze directions of the driver and the pedestrian. This thesis provides objective measurements of driver-pedestrian eye contact and demonstrates how eye contact may be detected and reconstructed for use in automated vehicles of the future.
both of which were synchronized. Participants’ locations were also tracked using a stereo camera equipped with pedestrian detection capabilities. Pedestrians imagined that they were on a real road and performed six types of trials
where they stood on / crossed from the left / right side curb in front of the stationary vehicle while either making eye contact or not making eye contact with the driver. The order of the trials was randomized, and each trial consisted of 3 repetitions of a driver-pedestrian interaction. If the driver and pedestrian were looking at each other at the same time there was eye contact, otherwise there was no eye contact. Significant differences in the percentages of eye contact between pedestrians standing on the left (median duration of 0.42 s) and the right (median duration of 0.54 s) were found. No significant differences in the percentages of eye contact between pedestrians crossing from the left (median duration of 1.23 s) and the right (median duration of 1.39 s) were found. Eye contact instants within trials were algorithmically detected by finding the angle between the 3-D gaze direction vectors of the driver and the pedestrian, and comparing it to an ‘eye contact threshold’. Trials were classified as either involving eye contact or not involving eye contact based on their percentages of eye contact instants. The classification performance of the algorithm was quantified using two ground truths: (1) Imposed eye contact (in half of the trials, participants were instructed to make eye contact; in the other half, participants were instructed not to make eye contact), and (2) Manually annotated areas of interest (AOIs) from the Tobii Pro Glasses 2 showing pedestrian eye contact seeking. The algorithm’s performance was found to be fair/poor and eye contact could be detected with an accuracy of 15-30°. A 3-D reconstruction of the driver-pedestrian interaction was achieved (in the form of an animation) by using the locations, head orientations and gaze directions of the driver and the pedestrian. This thesis provides objective measurements of driver-pedestrian eye contact and demonstrates how eye contact may be detected and reconstructed for use in automated vehicles of the future.
Advanced Driving Assistance Systems (ADAS) technologies like Adaptive Cruise Control (ACC) are becoming the normality for many users, and many major car manufacturers are introducing SAE level 2 and 3 automation systems into the market. The main advantage of Automated Vehicles (AV) will be the significant decrease in road accidents and casualties. However, a significant shift from conventional to automated vehicles must occur before it can have a positive impact on society. If the behaviour of the vehicle is not perceived as natural, the user will most likely not activate the ADAS features again. During this study a naturalistic dataset is used to investigate the driver behaviour, in the hope of bringing the current ACC logic to a more human-like behaviour that will feel more natural to the driver. The research question summarizes the final objective of this study: How can Naturalistic Driving Study (NDS) datasets be used in target performance setting for ACC systems? This study will answer the research question by studying human behaviour in the scene of following an accelerating vehicle. The main body of this thesis is divided in three chapters, one for each step of the research. First the information about the used datasets are provided together with the methodologies used to extract the relevant time-series data. Secondly driver behaviour models are created in order to mathematically characterize human behaviour. The strength of the created models is their ability to represent the full range of driver behaviour in terms of driving style. The aggressiveness parameter of the model can be easily adjusted to represent different percentiles of driver behaviour. This allows for a quick and effective tuning process: by changing a single parameter the driving style of the model can be fully modified. Finally, the driver behaviour models are implemented into a simulation environment. The models are simulated against an existing ACC logic in order to assess the difference in behaviour. The comparison highlighted two conclusions: first, the ACC logic behaves in a very conservative way compared to driver behaviour, especially when starting from standstill. Secondly, the kept by the ACC logic was not consistent throughout the speed range. This variation of the logic's driving style could result even more bothersome to the customer than its general conservative behaviour. The string stability of the driver behaviour models was also assessed. Although the proposed logic proved more stable than the regular ACC logic, it still cannot reach full string stability.
Hopefully, with the method developed in this study, the process of getting accustomed to this new technology will become easier for the customer. Thanks to the driver behaviour models the motion of the vehicle can feel familiar and predictable, with the controller becoming part of the Human Machine Interface (HMI). As the customer gets more familiar with this technology his expectation will also increase and change, especially as the levels of automation start to increase. This will inevitably push automakers to continue to improve the technology to deliver increasingly advanced and safe vehicles. ...
Hopefully, with the method developed in this study, the process of getting accustomed to this new technology will become easier for the customer. Thanks to the driver behaviour models the motion of the vehicle can feel familiar and predictable, with the controller becoming part of the Human Machine Interface (HMI). As the customer gets more familiar with this technology his expectation will also increase and change, especially as the levels of automation start to increase. This will inevitably push automakers to continue to improve the technology to deliver increasingly advanced and safe vehicles. ...
Advanced Driving Assistance Systems (ADAS) technologies like Adaptive Cruise Control (ACC) are becoming the normality for many users, and many major car manufacturers are introducing SAE level 2 and 3 automation systems into the market. The main advantage of Automated Vehicles (AV) will be the significant decrease in road accidents and casualties. However, a significant shift from conventional to automated vehicles must occur before it can have a positive impact on society. If the behaviour of the vehicle is not perceived as natural, the user will most likely not activate the ADAS features again. During this study a naturalistic dataset is used to investigate the driver behaviour, in the hope of bringing the current ACC logic to a more human-like behaviour that will feel more natural to the driver. The research question summarizes the final objective of this study: How can Naturalistic Driving Study (NDS) datasets be used in target performance setting for ACC systems? This study will answer the research question by studying human behaviour in the scene of following an accelerating vehicle. The main body of this thesis is divided in three chapters, one for each step of the research. First the information about the used datasets are provided together with the methodologies used to extract the relevant time-series data. Secondly driver behaviour models are created in order to mathematically characterize human behaviour. The strength of the created models is their ability to represent the full range of driver behaviour in terms of driving style. The aggressiveness parameter of the model can be easily adjusted to represent different percentiles of driver behaviour. This allows for a quick and effective tuning process: by changing a single parameter the driving style of the model can be fully modified. Finally, the driver behaviour models are implemented into a simulation environment. The models are simulated against an existing ACC logic in order to assess the difference in behaviour. The comparison highlighted two conclusions: first, the ACC logic behaves in a very conservative way compared to driver behaviour, especially when starting from standstill. Secondly, the kept by the ACC logic was not consistent throughout the speed range. This variation of the logic's driving style could result even more bothersome to the customer than its general conservative behaviour. The string stability of the driver behaviour models was also assessed. Although the proposed logic proved more stable than the regular ACC logic, it still cannot reach full string stability.
Hopefully, with the method developed in this study, the process of getting accustomed to this new technology will become easier for the customer. Thanks to the driver behaviour models the motion of the vehicle can feel familiar and predictable, with the controller becoming part of the Human Machine Interface (HMI). As the customer gets more familiar with this technology his expectation will also increase and change, especially as the levels of automation start to increase. This will inevitably push automakers to continue to improve the technology to deliver increasingly advanced and safe vehicles.
Hopefully, with the method developed in this study, the process of getting accustomed to this new technology will become easier for the customer. Thanks to the driver behaviour models the motion of the vehicle can feel familiar and predictable, with the controller becoming part of the Human Machine Interface (HMI). As the customer gets more familiar with this technology his expectation will also increase and change, especially as the levels of automation start to increase. This will inevitably push automakers to continue to improve the technology to deliver increasingly advanced and safe vehicles.