PANDA: Parkinson's Assistance and Notification Driving Aid

TL;DR

PANDA addresses the safety and independence challenges of driving with Parkinson's disease by delivering real-time, multi-modal in-car monitoring and timely alerts. The approach combines participatory design, driving-simulation data collection, and a data-driven pipeline to detect irregular PD-driving patterns via eye-tracking, steering, and pedal signals, with a privacy-conscious alert design. Key contributions include a PD-focused driving dataset, a real-time driving state monitor, and a user-centered evaluation that informs alert modalities and scenarios. The work shows PANDA's feasibility and provides a foundation for real-world deployment and broader applicability to driving disabilities beyond PD.

Abstract

Parkinson's Disease (PD) significantly impacts driving abilities, often leading to early driving cessation or accidents due to reduced motor control and increasing reaction times. To diminish the impact of these symptoms, we developed PANDA (Parkinson's Assistance and Notification Driving Aid), a multi-modality real-time alert system designed to monitor driving patterns continuously and provide immediate alerts for irregular driving behaviors, enhancing driver safety of individuals with PD. The system was developed through a participatory design process with 9 people with PD and 13 non-PD individuals using a driving simulator, which allowed us to identify critical design characteristics and collect detailed data on driving behavior. A user study involving individuals with PD evaluated the effectiveness of PANDA, exploring optimal strategies for delivering alerts and ensuring they are timely and helpful. Our findings demonstrate that PANDA has the potential to enhance the driving safety of individuals with PD, offering a valuable tool for maintaining independence and confidence behind the wheel.

Figures (13)

• Figure 1: Our study employs a user-centered, three-step design process: In the preliminary study, we gathered challenges and system design principles through interviews. During the system design stage, we collected user operation data and used it to train a PD driving state monitor system. In the user study phase, we obtained feedback on the monitor system with various alerts and derived effective alert strategies for drivers with PD.
• Figure 2: From interviews of people with PD and PD specialists and existing literature, we identify six critical design principles. This figure provides an overview of these principles.
• Figure 3: We set up two driving environments to collect real-time driving data from people with PD and non-PD participants: a city scene (shown on the left) and a highway scene (shown on the right).
• Figure 4: This chart compares the differences in the steering wheel, throttle, and brake data during straight-line driving among four participants: PD1, PD2, PD3, and NC1. (Data for all participants can be found in the Appendix.) In the chart, the horizontal axis represents time (in seconds), and the vertical axis represents steering wheel angle, throttle depth, and brake depth. The zero point on the vertical axis (center of the y-axis) indicates a steering wheel angle of 0, with positive values representing a right turn and negative values representing a left turn. For the throttle and brake, the highest values (at the top of the y-axis) represent fully released pedals, while lower values indicate deeper pedal engagement.
• Figure 5: We investigated the effectiveness of visual alerts across three dimensions: position, information presentation form, and alert body part. The image illustrates our design setting: for the position, we evaluated the HUD, dashboard, and car center control screen from left to right in (a); for information presentation form, we used red warning triangles with graphic symbols, text only, and red warning triangles with text, from left to right in (b); and for alert content, we focused on hands, feet, and eyes, from left to right in (c).