Automated vehicles promise a future where drivers can engage in non-driving tasks without hands on the steering wheels for a prolonged period. Nevertheless, automated vehicles may still need to occasionally hand the control back to drivers due to technology limitations and legal requirements. While some systems determine the need for driver takeover using driver context and road condition to initiate a takeover request, studies show that the driver may not react to it. We present DeepTake, a novel deep neural network-based framework that predicts multiple aspects of takeover behavior to ensure that the driver is able to safely take over the control when engaged in non-driving tasks. Using features from vehicle data, driver biometrics, and subjective measurements, DeepTake predicts the driver’s intention, time, and quality of takeover. We evaluate DeepTake performance using multiple evaluation metrics. Results show that DeepTake reliably predicts the takeover intention, time, and quality, with an accuracy of 96%, 93%, and 83%, respectively. Results also indicate that DeepTake outperforms previous state-of-the-art methods on predicting driver takeover time and quality. Our findings have implications for the algorithm development of driver monitoring and state detection.

Inspired by human visual attention, we propose a novel inverse reinforcement learning formulation using Maximum Entropy Deep Inverse Reinforcement Learning (MEDIRL) for predicting the visual attention of drivers in accident-prone situations. MEDIRL predicts fixation locations that lead to maximal rewards by learning a task-sensitive reward function from eye fixation patterns recorded from attentive drivers. Additionally, we introduce EyeCar, a new driver attention dataset in accident-prone situations. We conduct comprehensive experiments to evaluate our proposed model on three common benchmarks: (DR(eye)VE, BDD-A, DADA-2000), and our EyeCar dataset. Results indicate that MEDIRL outperforms existing models for predicting attention and achieves state-of-the-art performance. We present extensive ablation studies to provide more insights into different features of our proposed model.1

Over 35% of the world's population uses social media. Platforms like Facebook, Twitter, and Instagram have radically influenced the way individuals interact and communicate. These platforms facilitate both public and private communication with strangers and friends alike, providing rich insight into an individual's personality, health, and wellbeing. To date, many researchers have employed a variety of methods for extracting mental health-centric features from digital text communication (DTC) data, including natural language processing, social network analysis, and extraction of temporal discourse patterns. However, none have explored a hierarchical framework for extracting features from private messages with the goal of unifying approaches across methodological domains. Furthermore, while analyses of large, public corpora abound in existing literature, limited work has been done to explore the relationship between of private textual communications, personality traits, and symptoms of mental illness. We present a framework for constructing rich feature spaces from digital text communications. We then demonstrate the efficacy of our framework by applying it to a dataset of private Facebook messages in a college student population (N=103). Our results reveal key individual differences in temporal and relational behaviors, as well as language usage in relation to validated measures of trait-level anxiety, loneliness, and personality. This work represents a critical step forward in linking features of private social media messages to validated measures of mental health, wellbeing, and personality.