Concepts for Steering Wheel Hands-offdetection

Abstract

The growing deployment of Level 2 and Level 3 automated driving systems in automotive applications, particularly in passenger cars and commercial vehicles, has increased the importance of reliable driver monitoring. This is especially relevant in situations where the driver must remain engaged and be ready to retake control when necessary. In this context, steering wheel hands-on/hands-off detection (HOD) plays a key role in determining whether the driver maintains physical contact with the steering wheel. Currently, many HOD systems are based on capacitive sensing integrated into the steering wheel. While this approach is widely adopted in current advanced driver assistance systems (ADAS), it is associated with several practical limitations, including sensitivity to environmental influences, reduced performance when gloves are worn, dependence on hand placement, and challenges in achieving consistent robustness under real driving conditions. To address these limitations, this thesis explores alternative concepts for steering wheel hands-off detection by examining sensing technologies used in related engineering domains. A cross-domain review is conducted covering robotics, industrial automation, medical devices, and aviation, with the aim of identifying sensing principles that may be transferable to automotive applications. The selected technologies are assessed using criteria such as detection reliability, feasibility of integration, safety relevance, environmental robustness, cost, and compatibility with automotive requirements. To support the concept evaluation, an exploratory Proof of Concept (PoC) was developed to demonstrate selected sensing principles and assess their practical feasibility in a steering wheel context. The study highlights several promising sensing approaches that could either replace or complement existing capacitive systems. By providing a structured comparison of cross-domain technologies and an evaluation of their suitability for automotive use, the thesis offers a foundation for future development of more reliable HOD solutions. Ultimately, the work contributes to improved vehicle safety, more dependable ADAS functionality, and the continued advancement of human–machine interaction in automated driving.