Personalizing Driver Safety Interfaces via Driver Cognitive Factors Inference

• DOI: 10.48550/arxiv.2402.05893
• Publication Date: 2024-02-08
• Authors: Emily Sumner; Jonathan DeCastro; Jean Costa; Deepak Gopinath; Everlyne Kimani; Shabnam Hakimi; Allison Morgan; Andrew Best; Hieu Duy Nguyen; Daniel J. Brooks; Bassam ul Haq; Andrew Patrikalakis; Hiroshi Yasuda; Kate Sieck; Avinash Balachandran; Tiffany Chen; Guy Rosman

Recent advances in AI and intelligent vehicle technology hold promise to revolutionize mobility transportation, the form of advanced driving assistance (ADAS) interfaces. Although it is widely recognized that certain cognitive factors, such as impulsivity inhibitory control, are related risky behavior, play a significant role on-road risk-taking, existing systems fail leverage factors. Varying levels these factors could influence effectiveness acceptance driver safety We demonstrate an approach for personalizing interaction via interfaces triggered based on learned recurrent neural network. The network trained from population human drivers infer control recent behavior. Using high-fidelity motion simulator, we ability deduce then use inferred make instantaneous determinations whether or not engage interface. This interface aims decrease driver's speed during yellow lights reduce their inclination run through them.