Andreas Hula
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Andreas Hula has a total of 84 co-authors. Here are the top co-authors by number of shared publications:
- Peter Saleh
Austrian Institute of Technology
7 shared publications - Paul Rosenkranz
Austrian Institute of Technology
6 shared publications - Pendleton Read Montague
Fralin Biomedical Research Institute, Virginia Tech; Roanoke, VA, 24016, USA
4 shared publications - Christian Klösch
Austrian Institute of Technology
3 shared publications - Ángel Losada
Department of Accidentology, University Institute for Automobile Research Francisco Aparicio Izquierdo (INSIA-UPM), Universidad Politécnica de Madrid, 28031 Madrid, Spain
3 shared publications Andrea Schaub
AIT Austrian Institute of Technology
2 shared publications
Latest Publications
- Open Access
- 31 October 2026
- Published by Springer Nature
- p. 602-608
Identification of Mobility Challenges in European Cities Using Image-Based Sensors
To achieve carbon neutrality, cities must adopt new and innovative energy models for urban mobility, increasingly focusing on zero-emission and active mobility modes. The implementation of sustainable mobility solutions relies on their inclusivity, affordability and safety, as well as their consistency with users’ needs. In the Horizon Europe-funded project AMIGOS (Active Mobility Innovations for Green and Safe City Solutions), image-based sensors are utilized to identify and better understand current traffic and safety-related challenges in urban areas of nine european cities. The Mobility Observation Box (MOB) automatically detects and tracks traffic participants (pedestrians, bicyclists, cars, trucks, buses etc.) and their behavior in public space. Using surrogate safety measures such as TTC (Time-To-Collision) and PET (Post-Encroachment Time) and machine learning techniques, interactions and conflicts between motorized road users, pedestrians and cyclists are identified and analyzed, as well as their utilization of public urban space in real-time. This paper showcases the data collection, processing and analysis of an urban area in the municipality of Las Rozas, Spain. The results serve as a basis for the co-development of adapted mobility solutions, towards reducing traffic, increasing public and inclusive mobility modes and improving safety and co-habitation between different mobilities.
- 31 October 2026
- Published by Springer Nature
- p. 602-608
Open AccessIdentification of Mobility Challenges in European Cities Using Image-Based Sensors
To achieve carbon neutrality, cities must adopt new and innovative energy models for urban mobility, increasingly focusing on zero-emission and active mobility modes. The implementation of sustainable mobility solutions relies on their inclusivity, affordability and safety, as well as their consistency with users’ needs. In the Horizon Europe-funded project AMIGOS (Active Mobility Innovations for Green and Safe City Solutions), image-based sensors are utilized to identify and better understand current traffic and safety-related challenges in urban areas of nine european cities. The Mobility Observation Box (MOB) automatically detects and tracks traffic participants (pedestrians, bicyclists, cars, trucks, buses etc.) and their behavior in public space. Using surrogate safety measures such as TTC (Time-To-Collision) and PET (Post-Encroachment Time) and machine learning techniques, interactions and conflicts between motorized road users, pedestrians and cyclists are identified and analyzed, as well as their utilization of public urban space in real-time. This paper showcases the data collection, processing and analysis of an urban area in the municipality of Las Rozas, Spain. The results serve as a basis for the co-development of adapted mobility solutions, towards reducing traffic, increasing public and inclusive mobility modes and improving safety and co-habitation between different mobilities.
- Open Access
- 31 October 2026
- Published by Springer Nature
- p. 651-659
Safe Flows for Vulnerable Road Users in Urban Settings
Urban infrastructure often arises from centuries of building and adapting, sometimes struggling with the requirements of pedestrians, bicyclists and modern vehicle types or regulations. To ensure that all road users can interact safely in the public space formed thus, it is crucial that road users meet in a structured and safe manner. On a busy crossing in Vienna, Austria we identify potential safety issues between the flows of pedestrians, cyclists and motorized road users, by applying a machine learning methodology, based on available crash data and using clustering and supervised learning. This allows us to obtain an innovative data driven measure on the potential contributions of speed and interaction angle to investigating potential crash locations. Thus, this might be used for safety relevant adjustments to flows for all road users.
- 31 October 2026
- Published by Springer Nature
- p. 651-659
Open AccessSafe Flows for Vulnerable Road Users in Urban Settings
Urban infrastructure often arises from centuries of building and adapting, sometimes struggling with the requirements of pedestrians, bicyclists and modern vehicle types or regulations. To ensure that all road users can interact safely in the public space formed thus, it is crucial that road users meet in a structured and safe manner. On a busy crossing in Vienna, Austria we identify potential safety issues between the flows of pedestrians, cyclists and motorized road users, by applying a machine learning methodology, based on available crash data and using clustering and supervised learning. This allows us to obtain an innovative data driven measure on the potential contributions of speed and interaction angle to investigating potential crash locations. Thus, this might be used for safety relevant adjustments to flows for all road users.
- Open Access
- 29 October 2026
- Published by Springer Nature
- p. 421-426
Inconspicuous and Accurate Data Collection for Real-World Analysis of Motorized Vehicles Overtaking Bicycles on Secondary Rural Streets
Cycling offers an individual environmentally friendly mode of transportation readily available to many road users. Thus, with the goal of increasing road safety, the Austrian government has issued a law in 2022 that requires drivers of motorized vehicles to keep a lateral distance of at least 2 m to a bicycle when overtaking on rural roads. We propose an inconspicuous setup consisting of multiple sensor types, which can easily be mounted to different bicycles, to be as discreet as possible when measuring the lateral distance to overtaking vehicles. A method to automatically extract overtaking manoeuvres from the data is presented. On a secondary rural street in Lower Austria we measured lateral distances to vehicles that were overtaking a cyclist. The overtaking manoeuvres were automatically extracted from the data. Our analysis shows that 71% of drivers had a lateral distance of less than 2 m and another 18% were so close that the distance from the tip of the side mirror to the cyclist was most probably less than 2 m while overtaking. Based on findings from other studies as well as our work with the data collected we also discuss further improvements to be made to our setup.
- 29 October 2026
- Published by Springer Nature
- p. 421-426
Open AccessInconspicuous and Accurate Data Collection for Real-World Analysis of Motorized Vehicles Overtaking Bicycles on Secondary Rural Streets
Cycling offers an individual environmentally friendly mode of transportation readily available to many road users. Thus, with the goal of increasing road safety, the Austrian government has issued a law in 2022 that requires drivers of motorized vehicles to keep a lateral distance of at least 2 m to a bicycle when overtaking on rural roads. We propose an inconspicuous setup consisting of multiple sensor types, which can easily be mounted to different bicycles, to be as discreet as possible when measuring the lateral distance to overtaking vehicles. A method to automatically extract overtaking manoeuvres from the data is presented. On a secondary rural street in Lower Austria we measured lateral distances to vehicles that were overtaking a cyclist. The overtaking manoeuvres were automatically extracted from the data. Our analysis shows that 71% of drivers had a lateral distance of less than 2 m and another 18% were so close that the distance from the tip of the side mirror to the cyclist was most probably less than 2 m while overtaking. Based on findings from other studies as well as our work with the data collected we also discuss further improvements to be made to our setup.
- Open Access
- 20 October 2026
- Published by Elsevier in SSRN Electronic Journal
Prioritised Harmonised Infrastructure Support Schema for Cooperative, Connected and Automated Mobility (CCAM) and enabled Decision Making and Assessment Tools
Physical and Digital Infrastructure (PDI) is recognised to be of decisive importance for the enabling and massive deployment of Cooperative Connected and Automated Mobility (CCAM). Prior to the validation of its potential under different solution config-urations and scenarios to increase the Operational Design Domain (ODD) of Connected Automated Vehicles (CAVs), it is of out-most importance to agree on a clear and as holistic as possible definition and clustering of PDI for CCAM, assuming all different environmental contexts of deployment and their specific features – such as road geometry, environmental conditions or traffic conditions - as well as the capabilities and requirements together with potential assumptions on behalf of CAVs that will assume infrastructure support for their maximum possible benefit. In addition to assisting with the positioning and roadmap on technical ground, the consistent and unanimously accepted clustering of PDI and its support is vital for the informed and, thus hopefully, ”no regret” investments on infrastructure end and the coordinated interoperable roadmaps of the CCAM ecosystem, with the primary actors in this case being the Original Equipment Manufacturers (OEMs) and the Infrastructure Operators/ Owners. In this context, a meaningful parametric prioritization of the infrastructure support reflecting the needs of both types of actors and in alignment with their code of practice is a mandatory mechanism for exploiting the backbone knowledge towards the assessment of current read-iness level of the infrastructure, the decision making process on the context-specific cost-effective investments and, in turn, their impact on the infrastructure readiness to support and enable CCAM. The current manuscript describes the harmonized PDI Support Schema for CCAM that has been developed in the context of AUGMENTED CCAM Horizon Europe project, the prioritization principles and in turn mechanisms that have been applied on top of it, and, finally, the tools that have been developed to allow its maintenance (Improve the ACCAM PDI Schema Tool), and, via the exploitation of the prioritized schema semantics, its use for the assessment of infrastructure readiness for CCAM (ACCAM Infra-structure Readiness Assessment Tool) and for informed decision making in view of investments (CCAM Infrastructure Planning Tool). Whilst the schema, consisting of five (5) scaling Layers of Infrastructure support readiness for CCAM which classify a total of 81 physical and digital infrastructure elements, is an informative research resource that constitutes valuable knowledge for all CCAM value chain including the involved research and academia entities, its physical direct target user as well as the one of its enabled tools is the infrastructure operators and authorities of all types of traffic contexts (urban, peri-urban/rural, motorway).
- 20 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessPrioritised Harmonised Infrastructure Support Schema for Cooperative, Connected and Automated Mobility (CCAM) and enabled Decision Making and Assessment Tools
Physical and Digital Infrastructure (PDI) is recognised to be of decisive importance for the enabling and massive deployment of Cooperative Connected and Automated Mobility (CCAM). Prior to the validation of its potential under different solution config-urations and scenarios to increase the Operational Design Domain (ODD) of Connected Automated Vehicles (CAVs), it is of out-most importance to agree on a clear and as holistic as possible definition and clustering of PDI for CCAM, assuming all different environmental contexts of deployment and their specific features – such as road geometry, environmental conditions or traffic conditions - as well as the capabilities and requirements together with potential assumptions on behalf of CAVs that will assume infrastructure support for their maximum possible benefit. In addition to assisting with the positioning and roadmap on technical ground, the consistent and unanimously accepted clustering of PDI and its support is vital for the informed and, thus hopefully, ”no regret” investments on infrastructure end and the coordinated interoperable roadmaps of the CCAM ecosystem, with the primary actors in this case being the Original Equipment Manufacturers (OEMs) and the Infrastructure Operators/ Owners. In this context, a meaningful parametric prioritization of the infrastructure support reflecting the needs of both types of actors and in alignment with their code of practice is a mandatory mechanism for exploiting the backbone knowledge towards the assessment of current read-iness level of the infrastructure, the decision making process on the context-specific cost-effective investments and, in turn, their impact on the infrastructure readiness to support and enable CCAM. The current manuscript describes the harmonized PDI Support Schema for CCAM that has been developed in the context of AUGMENTED CCAM Horizon Europe project, the prioritization principles and in turn mechanisms that have been applied on top of it, and, finally, the tools that have been developed to allow its maintenance (Improve the ACCAM PDI Schema Tool), and, via the exploitation of the prioritized schema semantics, its use for the assessment of infrastructure readiness for CCAM (ACCAM Infra-structure Readiness Assessment Tool) and for informed decision making in view of investments (CCAM Infrastructure Planning Tool). Whilst the schema, consisting of five (5) scaling Layers of Infrastructure support readiness for CCAM which classify a total of 81 physical and digital infrastructure elements, is an informative research resource that constitutes valuable knowledge for all CCAM value chain including the involved research and academia entities, its physical direct target user as well as the one of its enabled tools is the infrastructure operators and authorities of all types of traffic contexts (urban, peri-urban/rural, motorway).
- Open Access
- 4 October 2025
- Published by Springer Nature
- p. 38-44
Automatic Detection of Asphalt Road Distress via Semantic Segmentation
We address the problem of automating the previously manual inspection of asphalt distress and damage, which is time-consuming and subjective. With the help of a semantic segmentation neural network, we automate this process and make a pixel perfect detection. Our method is based on a previously published approach, validated for detecting distress symptoms in asphalt surfaces, for highway road data. The main challenge with the previous approach was, that the original training material did not yet contain significant samples of classes like foliage, drainage and additional types of road surface damage relevant to smaller secondary roads. Upon trying to retrain the underlying neural network it was discovered that methodological improvements were necessary to accurately detect these new classes. For now, we tried a new combination of a momentum-based Adam-Optimizer and a Cosine based learning Rate scheduler with restarting epoch also called Cosine Annealing. In theory, this should evade local minima of the error function, compared to the previously used polynomial decay without restarting epoch. The combination of both should yield a better minimum of the error function and higher accuracy rate.
- 4 October 2025
- Published by Springer Nature
- p. 38-44
Open AccessAutomatic Detection of Asphalt Road Distress via Semantic Segmentation
We address the problem of automating the previously manual inspection of asphalt distress and damage, which is time-consuming and subjective. With the help of a semantic segmentation neural network, we automate this process and make a pixel perfect detection. Our method is based on a previously published approach, validated for detecting distress symptoms in asphalt surfaces, for highway road data. The main challenge with the previous approach was, that the original training material did not yet contain significant samples of classes like foliage, drainage and additional types of road surface damage relevant to smaller secondary roads. Upon trying to retrain the underlying neural network it was discovered that methodological improvements were necessary to accurately detect these new classes. For now, we tried a new combination of a momentum-based Adam-Optimizer and a Cosine based learning Rate scheduler with restarting epoch also called Cosine Annealing. In theory, this should evade local minima of the error function, compared to the previously used polynomial decay without restarting epoch. The combination of both should yield a better minimum of the error function and higher accuracy rate.
- Open Access
- 3 September 2025
- Published by Springer Nature
- p. 3-8
Extracting Parameters of Vulnerable Road Users’ Behavior from Video Observation for Modeling of Urban Microsimulation
Microsimulation is an essential tool in the planning and evaluation of large infrastructure projects or the implementation of new traffic rules for the inclusion of new road users. Efficient microsimulation on a citywide scale is provided by several software solutions, however, the representation of behaviors of unmotorized vulnerable road users (VRUs i.e., pedestrians, cyclists, scooter users) with their many degrees of freedom (including suddenly stepping on the road, crossing red lights) and their close interactions due to lack of space on sidewalks and other areas tend to be highly simplified. This work utilizes camera data collected in a busy road environment with several close interactions of all types of urban road users, to define decision-making rules and models of vulnerable road user behavior to be used in microsimulation. The simulation models are based on the observed interaction statistics, through the use of Surrogate Safety Measures (SSM, in particular Time-to-Collision-(TTC-), Post Encroachment Time- (PET-)), and the assessment of the kinematic behavior of interacting users in potential traffic conflicts.
- 3 September 2025
- Published by Springer Nature
- p. 3-8
Open AccessExtracting Parameters of Vulnerable Road Users’ Behavior from Video Observation for Modeling of Urban Microsimulation
Microsimulation is an essential tool in the planning and evaluation of large infrastructure projects or the implementation of new traffic rules for the inclusion of new road users. Efficient microsimulation on a citywide scale is provided by several software solutions, however, the representation of behaviors of unmotorized vulnerable road users (VRUs i.e., pedestrians, cyclists, scooter users) with their many degrees of freedom (including suddenly stepping on the road, crossing red lights) and their close interactions due to lack of space on sidewalks and other areas tend to be highly simplified. This work utilizes camera data collected in a busy road environment with several close interactions of all types of urban road users, to define decision-making rules and models of vulnerable road user behavior to be used in microsimulation. The simulation models are based on the observed interaction statistics, through the use of Surrogate Safety Measures (SSM, in particular Time-to-Collision-(TTC-), Post Encroachment Time- (PET-)), and the assessment of the kinematic behavior of interacting users in potential traffic conflicts.
- Open Access
- 3 September 2025
- Published by Springer Nature
- p. 90-96
Safety Relevant Behaviour-Based Driving Dynamics Indicators of E-Scooter Usage in Parcours Situations
E-Scooters provide a fast, easily available, easy to use means of transportation for a broad range of users in urban traffic. Apart from adjustments of rules and policies of E-scooter use, increasing safety for these vulnerable road users is challenging, given driven speeds, handling of e-scooters and the lack of physical protection for E-scooter riders. One approach to understanding the needs of E-scooter users is to study their driving behaviours in everyday and specific challenging situations. The project SEED (Safe E-Scooter Driving) aims to provide and discuss such data, which was collected from more than 80 subjects during an extensive three-day test trial. Three different E-scooters have been used, each equipped with measurement devices to precisely collect data of their motion states. Voluntary test subjects were instructed to pass a diverse driving parcours, including braking manoeuvres. We study the effects of different scooter types on manoeuvrability (indicated by measures such as standard deviation of steering speed and roll-rate or decelerations realized in real conditions), whilst correcting for influences of age, sex and several other parameters on measured driving behaviours. This yields conclusions on what E-Scooter types require which behavioural adjustments to drive safely.
- 3 September 2025
- Published by Springer Nature
- p. 90-96
Open AccessSafety Relevant Behaviour-Based Driving Dynamics Indicators of E-Scooter Usage in Parcours Situations
E-Scooters provide a fast, easily available, easy to use means of transportation for a broad range of users in urban traffic. Apart from adjustments of rules and policies of E-scooter use, increasing safety for these vulnerable road users is challenging, given driven speeds, handling of e-scooters and the lack of physical protection for E-scooter riders. One approach to understanding the needs of E-scooter users is to study their driving behaviours in everyday and specific challenging situations. The project SEED (Safe E-Scooter Driving) aims to provide and discuss such data, which was collected from more than 80 subjects during an extensive three-day test trial. Three different E-scooters have been used, each equipped with measurement devices to precisely collect data of their motion states. Voluntary test subjects were instructed to pass a diverse driving parcours, including braking manoeuvres. We study the effects of different scooter types on manoeuvrability (indicated by measures such as standard deviation of steering speed and roll-rate or decelerations realized in real conditions), whilst correcting for influences of age, sex and several other parameters on measured driving behaviours. This yields conclusions on what E-Scooter types require which behavioural adjustments to drive safely.
- Open Access
- 4 July 2025
- Published by Elsevier in Accident Analysis & Prevention
- Vol. 220, 108155
Trajectory-based indicators to determine the local character of intersection conflicts: A micro-spatial analysis
- 4 July 2025
- Published by Elsevier in Accident Analysis & Prevention
- Vol. 220, 108155
Open AccessTrajectory-based indicators to determine the local character of intersection conflicts: A micro-spatial analysis
- Open Access
- 7 March 2025
- Published by Springer Nature
- p. 127-133
Using a Motorcycle Probe Vehicle to Provide Infrastructure Information for Powered Two Wheelers
Powered two wheelers are a consistently popular mode of transportation and riding them is a widely practiced recreational activity. However, with regards to safety powered two wheelers are clearly vulnerable road users, for whom accidents of all types yield more severe outcomes on average than for larger vehicles. One way to ensure the safety of powered two-wheeler riders is to identify challenging infrastructure properties, such as transversal evenness qualities or potholes, through the use of probe vehicles. We present the analysis of infrastructure properties through data collected with a motorcycle probe vehicle, previously employed to study human driving dynamics and assess the risk thereof. We present the first steps towards a new standardized evaluation of motorcycle driving dynamics data for this purpose and show that our outcomes can be achieved based on in-vehicle driving dynamics data. This holds the potential to enable the provision of safety relevant data from everyday driven vehicles, which represents the needs of powered two wheelers as much as those of passenger cars or larger vehicle-types and could serve as a template for similar analyses employing smaller probe vehicles like bicycles or (e-)scooters.
- 7 March 2025
- Published by Springer Nature
- p. 127-133
Open AccessUsing a Motorcycle Probe Vehicle to Provide Infrastructure Information for Powered Two Wheelers
Powered two wheelers are a consistently popular mode of transportation and riding them is a widely practiced recreational activity. However, with regards to safety powered two wheelers are clearly vulnerable road users, for whom accidents of all types yield more severe outcomes on average than for larger vehicles. One way to ensure the safety of powered two-wheeler riders is to identify challenging infrastructure properties, such as transversal evenness qualities or potholes, through the use of probe vehicles. We present the analysis of infrastructure properties through data collected with a motorcycle probe vehicle, previously employed to study human driving dynamics and assess the risk thereof. We present the first steps towards a new standardized evaluation of motorcycle driving dynamics data for this purpose and show that our outcomes can be achieved based on in-vehicle driving dynamics data. This holds the potential to enable the provision of safety relevant data from everyday driven vehicles, which represents the needs of powered two wheelers as much as those of passenger cars or larger vehicle-types and could serve as a template for similar analyses employing smaller probe vehicles like bicycles or (e-)scooters.