Postdoc Data-driven Prediction of Human Body Motion in Automated Vehicles

The potential discomfort and motion sickness experienced by passengers during automated journeys pose a significant challenge to their widespread adoption. Envisaged automated vehicle (AV) designs and their driving behavior are expected to provoke motion sickness and discomfort, hindering passengers' ability to enjoy their commute time. Despite substantial investments in AV technology, the importance of motion comfort has been largely overlooked. Fundamental questions regarding occupants' postural control (when and how they are activated) in AVs remain unanswered, which has led to the lack of human body models (HBM) able to predict human motion and postural control (both anticipatory as preparation for upcoming perturbation, and compensatory for restoring balance after perturbation).
To model human motion, researchers have employed simplified efficient models, which are faster to run than complex human body models and useful for early-stage design evaluations. However, these simplified models may fail to capture the intricacies of the human body's response to motion. Advanced active human body models offer highly detailed information but require specialized expertise and significant computational time. Furthermore, neither type of model adequately captures occupants' anticipatory and compensatory postural control based on upcoming or experienced motion.
To that end, this project will explore:

1. How data driven control techniques can be employed to gain insight in occupants’ postural adjustments while being driven?
2. To what extend can we employ feedforward and feedback components to capture anticipation and compensation?

For this, we will leverage extensive data from motion capture systems, wearable devices, and other sources from a groundbreaking experiment and we will apply nonlinear learning control techniques to model the complexities of human movement and predict occupants’ postural control while being driven. These insights and models will enable the design of ergonomic solutions, optimization of occupant-vehicle interaction and significantly improved motion comfort in AVs. Only then, automated journeys will not only be safe and sustainable but also comfortable and enjoyable for all passengers.
The main deliverable of the project will be validated simplified human body models with data-driven control algorithms that capture occupants' postural adjustments. The models will be validated using existing data that the group has. The researcher is expected to lead the project and deliver publications (journal & conferences) that disseminate the work.
At TU Delft, you will be supervised by Dr. Georgios Papaioannou (Intelligent Vehicles Section at Cognitive Robotics Department) and Dr. Meichen Guo (Delft Center for Systems and Control). The project is funded as Cohesion grant, a internal TUD scheme to support tenure trackers. The Postdoc will be hosted by the Intelligent Vehicles Section. The section consists of three groups (Machine Perception, Human Factors and Dynamics & Control). The groups collaborate together to increase the safety, comfort and sustainability of transportation by means of automated driving. You will work be within the Human Factors group and collaborate with BSc, MSc, and PhD candidates in the involved research groups.
The Human Factors' PhD students and Postdocs cover a wide range of topics from understanding and modelling motion sickness in automated vehicles, biomecanics and postural control, countermeasures for motion sickness and discomfort, teleoperation, perceived safety in automated vehicles, and others. Opportunities for growing your career will be available, such as mentoring BSc/MSc Students, conference presentations, networking, and teaching.

Requirements:

The candidate shall hold a:

• PhD in Mechanical Engineering, Biomechanics, Systems and Control, or any comparable studies by the start date of the position.
• Strong scientific programming skills.
• Strong written and oral communication skills in English.

The following aspects will help you stand out:

• Knowledge of data-driven control algorithms, biomechanical modelling, system identification, machine learning, control theory.
• Prior experimental experience on human body dynamics and motion comfort.
• A strong academit track record with publications in the relevant topics.
• The ability to act independently as well as to collaborate effectively with members of a larger team.

Keep in mind that this describes the background we imagine would best fit the role. Even if you do not meet all of the requirements and feel that you are up for the task, we absolutely want to see your application!

Salary Benefits:

Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities (salary based on scale 10, indication: € 3.877 - € 5.090  per month gross). The TU Delft offers a customisable compensation package, discounts on health insurance, and a monthly work costs contribution. Flexible work schedules can be arranged.
For international applicants, TU Delft has the Coming to Delft Service. This service provides information for new international employees to help you prepare the relocation and to settle in the Netherlands. The Coming to Delft Service offers a Dual Career Programme for partners and they organise events to expand your (social) network.
This postdoc position has a fixed-term contract of 12 months.

32 - 40 hours per week

Mekelweg 2