Ifsttar PhD subject

French version
Detailed form :

Title : Comfortable seating postures and their monitoring in a highly automated vehicle

Main host Laboratory - Referent Advisor   -

Director of the main host Laboratory   -

PhD Speciality Mécanique

Axis of the performance contract 1 - COP2017 - Efficient transport and safe travel

Main location Bron

Doctoral affiliation UNIVERSITE CLAUDE-BERNARD-LYON 1

PhD school MEGA (MECANIQUE, ENERGETIQUE, GENIE CIVIL, ACOUSTIQUE)

Planned PhD supervisor WANG Xuguang  -  Université Gustave Eiffel  -  TS2 - LBMC

Planned financing Thèse CIFRE  - PSA Peugeot Citroën

Abstract

Résumé (10000 caractères maximum pour le portail des thèses):

Context
Technologies for automated driving are developing rapidly. Based on the recent report prepared for Governors Highway Safety Association by Hedlund (2017), partial self-driving (Level 2, by SAE) features are now available. Levels 3 (limited self-driving, but drivers have to be ready to take control when required) though 5 (full self-driving under all conditions) are being tested extensively. With increasing level of driving automatization, vehicle interior has to be adapted for new activities (such as reading books, playing smart phones, working with a computer, sleeping, etc…) that a driver may adopt when he/she is not required for driving. Meanwhile, though automated vehicles (AVs) are supposed to dramatically reduce the number of human factor related accidents, accidents are unavoidable due to other factors (Fagnant and Kockelman, 2015). An accurate monitoring of drivers’ posture can provide necessary inputs for optimizing protection systems as well as for detecting driver’s inattention (Dong et al., 2011). However, the interior of currently existing vehicles is designed so that drivers can adopt a comfortable posture mainly for driving. What are the new activities that a driver may adopt? What are the corresponding preferred seated postures? How the knowledge of comfortable postures corresponding to these new activities can be used for accurately monitoring drivers’ position? These are the main research questions that the present research project aims to answer.

Preferred/comfortable postures for driving have been largely investigated in the past (Schmidt et al., 2014) and even recently at our lab (Peng et al., 2017 and 2018). With the strong vision requirement (drivers have to keep their eyes on road) and position constraints on the hands (by the steering wheel) and feet (by the pedals), the range of driving postural variation is certainly small compared to the one in the future highly AVs, for which these constraints may be completely removed when doing other activities than driving. The interior design of future AVs has to take into account the requirement of a larger postural variation to accommodate new activities. For example, results of a recent on-line survey from 122 drivers by Yang et al. (2018) show that people may prefer a more reclined seating for having a rest and require more space around the torso and knee for other activities than driving in future AVs. There exist few quantitative investigations on comfortable postures for these new activities. Existing design aided tools such as digital human modeling packages (e.g. RAMSIS) largely used by car manufacturers are not able to predict these potential postures and to assess their discomfort.

Considerable researches have been initiated to develop sensors and methods to monitor the driver’s posture, based on computer vision (mono camera and stereovision) technologies, ultrasonic system, capacitive proximity sensors, pressure pads etc. To develop a cheap, non-intrusive and accurate postural monitoring system is still a challenging task. It is more likely that different sensors would be used providing redundant information about the position of drivers and other occupants in a vehicle. It is believed that a priori knowledge on body posture in a vehicle will reduce the amount of measurements by sensors for developing future driver’s postural monitoring system. Such a priori knowledge could be useful for correcting inconsistent reconstructed body posture in case of obstruction of some body parts when using an optical motion capture system. Recently, we implemented the motion graph algorithm proposed by Plantard et al. (2016) in computer graphics, collected a set of driving postures using an experimental mock-up and used them to correct inconsistent postured provided by one depth KINECT camera (Zhao et al, 2018). First results are encouraging and have helped in identifying research problems related to posture/motion indexing and classification for a more efficient reuse of data base.

Objective
Therefore, the main objective of this project is 1) to develop the knowledge on comfortable postures corresponding to new activities that a driver may adopt in a future AV and 2) to explore this a prior knowledge to help developing an in-vehicle occupant monitoring system.

Proposed approach
The present project will rely on high expertise of IFSTTAR (more specifically the research laboratory LBMC) in digital human modeling and its application for vehicle interior design (e.g. European Project DHErgo with LBMC as coordinator, Collaborative projects with car manufacturers Renault, PSA and Toyota Motor Europe) (see Peng et al, 2017, 2018 for more recent publications). LBMC has up-to-date motion capture and analysis systems and developed an advanced motion analysis and simulation tool using a digital human model (Wang et al, 2006, Monnier et al, 2009). More recently, LBMC built a new multi-adjustable experimental seat with 13 motorized adjustments (Beurier et al., 2017), making it possible to simulate a large range of sitting postures that might appear in future autonomous vehicles.
After a review of existing literature, an experimental study will be performed to collect the data of preferred postures corresponding to the most likely non-driving related activities using the new LBMC experimental seat from a sample of participants covering a large range of body size variation. A likely vehicle interior will be defined and used when collecting data. Statistical regressions will be developed allowing the prediction of preferred body posture for different activities in function of body size and environment constraints. The collected data will also be explored to help developing an in-vehicle occupant monitoring system using the postural monitoring algorithm developed by LBMC (Zhao et al., 2018).

Candidate profile
Master in mechanical engineering with experience in biomechanics, Good programming skills in Matlab and C++ , French speaking level required for everyday life

Supervision team
The work will be mainly supervised by Xuguang Wang et Georges Beurier from LBMC IFSTTAR and Dr Lisa Denninger from PSA.

References :
Beurier, G., Cardoso, M., and Wang, X., 2017. A new multi-adjustable experimental seat for investigating biomechanical factors of sitting discomfort. SAE Technical Paper 2017-01-1393, doi:10.4271/2017-01-1393
Chen J., Jiang B., Song S., Wang H. and X. Wang, 2016. In-Vehicle Driving Posture Reconstruction from 3D Scanning Data Using a 3D Digital Human Modeling Tool. SAE Technical Paper 2016-01-1357, 2016, doi:10.4271/2016-01-1357
Dong, Y., Z. Hu, K. Uchimura, and N. Murayama, 2011. Driver Inattention Monitoring System for Intelligent Vehicles: A Review. IEEE Transactions on Intelligent Transportation Systems 12, no. 2 (June 2011): 596–614. doi:10.1109/TITS.2010.2092770.
Fagnant, D.J., and K. Kockelman, 2015. Preparing a Nation for Autonomous Vehicles: Opportunities, Barriers and Policy Recommendations. Transportation Research Part A: Policy and Practice 77, no. Supplement C (July 1, 2015): 167–81. doi:10.1016/j.tra.2015.04.003.
Hedlund J. (2017). Autonomous Vehicles Meet Human Drivers: Traffic Safety Issues for States. http://www.ghsa.org/sites/default/files/2017-01/AV%202017%20-%20FINAL.pdf
Monnier G., Wang X., Trasbot J., 2009. RPx: A motion simulation tool for car interior design. In: Handbook of Digital Human Modeling: Research for Applied Ergonomics and Human Factors Engineering, ed. Vincent G. Duffy. CRC press Taylor and Francis Group. 2009
Plantard P, Shum H P H, Multon F., 2016. Filtered pose graph for efficient kinect pose reconstruction. Multimedia Tools and Applications, 2016: 1-22
Peng J., Wang X. and Denninger L. (2018) Effects of anthropometric variables and seat height on automobile drivers’ preferred posture with the presence of the clutch. Human factors, Vol. 60, No. 2, March 2018, pp. 172–190, DOI: 10.1177/0018720817741040
Peng J., Wang X. and Denninger L. (2017) Ranges of the least uncomfortable joint angles for assessing automotive driving posture. Applied Ergonomics, 61, 12-21, DOI.10.1016/j.apergo.2016.12.021
Schmidt, S., Amereller, M., Franz, M., Kaiser, R., and Schwirtz, A., 2014. A literature review on optimum and preferred joint angles in automotive sitting posture. Appl Ergon 45, 247–260.
Shu, C., Wuhrer, S. and Xi, P., 2012. 3D anthrhropometric data processing. Int. J. of Human Factors Modeling and Simulation, Vol.3; n°2, 133-146(2012)
Yang, Y., Klinkner, J., Bengler, J., 2018. How will the driver sit in an automated vehicle ? The qualitative and quantitative descriptions of non-driving postures (NDPs) when non-driving-related-tasks (NDRTs) are conducted., DOI: 10.1007/978-3-319-96074-6_44, Proceedings of the 20th Congress of the International Ergonomics Association (IEA 2018)
Zhao, M., Beurier, G., Wang, H., Wang, X., 2018. Driver posture monitoring in intelligent vehicles using a depth camera. 2018 SAE World Congress. Detroit, April 10-12, 2018. SAE Technical Paper 2018-01-0505, 2018, doi:10.4271/2018-01-0505

To apply: please send CV, motivation and recommendation letters to:
xuguang.wang@ifsttar.fr or lisa.denninger@mpsa.com

Keywords : /automated vehicule, posture, monitoring, comfort, human model

Main host Laboratory - Referent Advisor	-
Director of the main host Laboratory	-
PhD Speciality	Mécanique
Axis of the performance contract	1 - COP2017 - Efficient transport and safe travel
Main location	Bron
Doctoral affiliation	UNIVERSITE CLAUDE-BERNARD-LYON 1
PhD school	MEGA (MECANIQUE, ENERGETIQUE, GENIE CIVIL, ACOUSTIQUE)
Planned PhD supervisor	WANG Xuguang - Université Gustave Eiffel - TS2 - LBMC
Planned financing	Thèse CIFRE - PSA Peugeot Citroën
Abstract Résumé (10000 caractères maximum pour le portail des thèses): Context Technologies for automated driving are developing rapidly. Based on the recent report prepared for Governors Highway Safety Association by Hedlund (2017), partial self-driving (Level 2, by SAE) features are now available. Levels 3 (limited self-driving, but drivers have to be ready to take control when required) though 5 (full self-driving under all conditions) are being tested extensively. With increasing level of driving automatization, vehicle interior has to be adapted for new activities (such as reading books, playing smart phones, working with a computer, sleeping, etc…) that a driver may adopt when he/she is not required for driving. Meanwhile, though automated vehicles (AVs) are supposed to dramatically reduce the number of human factor related accidents, accidents are unavoidable due to other factors (Fagnant and Kockelman, 2015). An accurate monitoring of drivers’ posture can provide necessary inputs for optimizing protection systems as well as for detecting driver’s inattention (Dong et al., 2011). However, the interior of currently existing vehicles is designed so that drivers can adopt a comfortable posture mainly for driving. What are the new activities that a driver may adopt? What are the corresponding preferred seated postures? How the knowledge of comfortable postures corresponding to these new activities can be used for accurately monitoring drivers’ position? These are the main research questions that the present research project aims to answer. Preferred/comfortable postures for driving have been largely investigated in the past (Schmidt et al., 2014) and even recently at our lab (Peng et al., 2017 and 2018). With the strong vision requirement (drivers have to keep their eyes on road) and position constraints on the hands (by the steering wheel) and feet (by the pedals), the range of driving postural variation is certainly small compared to the one in the future highly AVs, for which these constraints may be completely removed when doing other activities than driving. The interior design of future AVs has to take into account the requirement of a larger postural variation to accommodate new activities. For example, results of a recent on-line survey from 122 drivers by Yang et al. (2018) show that people may prefer a more reclined seating for having a rest and require more space around the torso and knee for other activities than driving in future AVs. There exist few quantitative investigations on comfortable postures for these new activities. Existing design aided tools such as digital human modeling packages (e.g. RAMSIS) largely used by car manufacturers are not able to predict these potential postures and to assess their discomfort. Considerable researches have been initiated to develop sensors and methods to monitor the driver’s posture, based on computer vision (mono camera and stereovision) technologies, ultrasonic system, capacitive proximity sensors, pressure pads etc. To develop a cheap, non-intrusive and accurate postural monitoring system is still a challenging task. It is more likely that different sensors would be used providing redundant information about the position of drivers and other occupants in a vehicle. It is believed that a priori knowledge on body posture in a vehicle will reduce the amount of measurements by sensors for developing future driver’s postural monitoring system. Such a priori knowledge could be useful for correcting inconsistent reconstructed body posture in case of obstruction of some body parts when using an optical motion capture system. Recently, we implemented the motion graph algorithm proposed by Plantard et al. (2016) in computer graphics, collected a set of driving postures using an experimental mock-up and used them to correct inconsistent postured provided by one depth KINECT camera (Zhao et al, 2018). First results are encouraging and have helped in identifying research problems related to posture/motion indexing and classification for a more efficient reuse of data base. Objective Therefore, the main objective of this project is 1) to develop the knowledge on comfortable postures corresponding to new activities that a driver may adopt in a future AV and 2) to explore this a prior knowledge to help developing an in-vehicle occupant monitoring system. Proposed approach The present project will rely on high expertise of IFSTTAR (more specifically the research laboratory LBMC) in digital human modeling and its application for vehicle interior design (e.g. European Project DHErgo with LBMC as coordinator, Collaborative projects with car manufacturers Renault, PSA and Toyota Motor Europe) (see Peng et al, 2017, 2018 for more recent publications). LBMC has up-to-date motion capture and analysis systems and developed an advanced motion analysis and simulation tool using a digital human model (Wang et al, 2006, Monnier et al, 2009). More recently, LBMC built a new multi-adjustable experimental seat with 13 motorized adjustments (Beurier et al., 2017), making it possible to simulate a large range of sitting postures that might appear in future autonomous vehicles. After a review of existing literature, an experimental study will be performed to collect the data of preferred postures corresponding to the most likely non-driving related activities using the new LBMC experimental seat from a sample of participants covering a large range of body size variation. A likely vehicle interior will be defined and used when collecting data. Statistical regressions will be developed allowing the prediction of preferred body posture for different activities in function of body size and environment constraints. The collected data will also be explored to help developing an in-vehicle occupant monitoring system using the postural monitoring algorithm developed by LBMC (Zhao et al., 2018). Candidate profile Master in mechanical engineering with experience in biomechanics, Good programming skills in Matlab and C++ , French speaking level required for everyday life Supervision team The work will be mainly supervised by Xuguang Wang et Georges Beurier from LBMC IFSTTAR and Dr Lisa Denninger from PSA. References : Beurier, G., Cardoso, M., and Wang, X., 2017. A new multi-adjustable experimental seat for investigating biomechanical factors of sitting discomfort. SAE Technical Paper 2017-01-1393, doi:10.4271/2017-01-1393 Chen J., Jiang B., Song S., Wang H. and X. Wang, 2016. In-Vehicle Driving Posture Reconstruction from 3D Scanning Data Using a 3D Digital Human Modeling Tool. SAE Technical Paper 2016-01-1357, 2016, doi:10.4271/2016-01-1357 Dong, Y., Z. Hu, K. Uchimura, and N. Murayama, 2011. Driver Inattention Monitoring System for Intelligent Vehicles: A Review. IEEE Transactions on Intelligent Transportation Systems 12, no. 2 (June 2011): 596–614. doi:10.1109/TITS.2010.2092770. Fagnant, D.J., and K. Kockelman, 2015. Preparing a Nation for Autonomous Vehicles: Opportunities, Barriers and Policy Recommendations. Transportation Research Part A: Policy and Practice 77, no. Supplement C (July 1, 2015): 167–81. doi:10.1016/j.tra.2015.04.003. Hedlund J. (2017). Autonomous Vehicles Meet Human Drivers: Traffic Safety Issues for States. http://www.ghsa.org/sites/default/files/2017-01/AV%202017%20-%20FINAL.pdf Monnier G., Wang X., Trasbot J., 2009. RPx: A motion simulation tool for car interior design. In: Handbook of Digital Human Modeling: Research for Applied Ergonomics and Human Factors Engineering, ed. Vincent G. Duffy. CRC press Taylor and Francis Group. 2009 Plantard P, Shum H P H, Multon F., 2016. Filtered pose graph for efficient kinect pose reconstruction. Multimedia Tools and Applications, 2016: 1-22 Peng J., Wang X. and Denninger L. (2018) Effects of anthropometric variables and seat height on automobile drivers’ preferred posture with the presence of the clutch. Human factors, Vol. 60, No. 2, March 2018, pp. 172–190, DOI: 10.1177/0018720817741040 Peng J., Wang X. and Denninger L. (2017) Ranges of the least uncomfortable joint angles for assessing automotive driving posture. Applied Ergonomics, 61, 12-21, DOI.10.1016/j.apergo.2016.12.021 Schmidt, S., Amereller, M., Franz, M., Kaiser, R., and Schwirtz, A., 2014. A literature review on optimum and preferred joint angles in automotive sitting posture. Appl Ergon 45, 247–260. Shu, C., Wuhrer, S. and Xi, P., 2012. 3D anthrhropometric data processing. Int. J. of Human Factors Modeling and Simulation, Vol.3; n°2, 133-146(2012) Yang, Y., Klinkner, J., Bengler, J., 2018. How will the driver sit in an automated vehicle ? The qualitative and quantitative descriptions of non-driving postures (NDPs) when non-driving-related-tasks (NDRTs) are conducted., DOI: 10.1007/978-3-319-96074-6_44, Proceedings of the 20th Congress of the International Ergonomics Association (IEA 2018) Zhao, M., Beurier, G., Wang, H., Wang, X., 2018. Driver posture monitoring in intelligent vehicles using a depth camera. 2018 SAE World Congress. Detroit, April 10-12, 2018. SAE Technical Paper 2018-01-0505, 2018, doi:10.4271/2018-01-0505 To apply: please send CV, motivation and recommendation letters to: xuguang.wang@ifsttar.fr or lisa.denninger@mpsa.com
Keywords :	/automated vehicule, posture, monitoring, comfort, human model

List of topics

Applications closed