Lateral Evasive Maneuver with Shared Control Algorithm: A Simulator Study

• Published in: Sensors (Basel, Switzerland) Vol. 24; no. 2; p. 562
• Main Authors: Sarabia, Joseba, Marcano, Mauricio, Díaz, Sergio, Zubizarreta, Asier, Pérez, Joshué
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.01.2024; MDPI
• Subjects: Aggression; Algorithms; Analysis; automated driving; Automation; Autonomous Vehicles; Control algorithms; driver-automation cooperation; Haptics; Humans; Motorcycles; Roads & highways; safety; shared control; simulator-based study; user acceptance; Vehicles; simulator-based study; user acceptance; safety; automated driving; driver-automation cooperation; shared control
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s24020562

Shared control algorithms have emerged as a promising approach for enabling real-time driver automated system cooperation in automated vehicles. These algorithms allow human drivers to actively participate in the driving process while receiving continuous assistance from the automated system in specific scenarios. However, despite the theoretical benefits being analyzed in various works, further demonstrations of the effectiveness and user acceptance of these approaches in real-world scenarios are required due to the involvement of the human driver in the control loop. Given this perspective, this paper presents and analyzes the results of a simulator-based study conducted to evaluate a shared control algorithm for a critical lateral maneuver. The maneuver involves the automated system helping to avoid an oncoming motorcycle that enters the vehicle's lane. The study's goal is to assess the algorithm's performance, safety, and user acceptance within this specific scenario. For this purpose, objective measures, such as collision avoidance and lane departure prevention, as well as subjective measures related to the driver's sense of safety and comfort are studied. In addition, three levels of assistance (gentle, intermediate, and aggressive) are tested in two driver state conditions (focused and distracted). The findings have important implications for the development and execution of shared control algorithms, paving the way for their incorporation into actual vehicles.