Sion Pizzi
Pronouns: He/Him/His
UROP Fellowship: Community College Summer Fellowship Program
Delta College
Research Mentor(s): Neda Masoud, PhD
Civil and Environmental Engineering
Presentation Date: Thursday, July 30, 2020 | Session 3 | Presenter: 6
Authors: Sion Pizzi, Ethan Zhang, Rajesh Malhan, Neda Masoud
Abstract
Connected vehicle (CV) technology is being developed under the premise that it will increase safety, enhance mobility, and curb the environmental footprint of the transportation sector. However, for these benefits to be realized in reality, data-driven applications need to be developed. As the world becomes more automated, we need machines to increasingly keep us safe as well. We have several safety precautions for drivers inside a car, such as seatbelts and airbags. This project focuses on safety of pedestrians: the most vulnerable road users. The goal of this study is to predict and prevent scenarios that pose safety risks to pedestrians at intersections.
In order to develop our application, we need to gather data in the form of pedestrian trajectories in the vicinity of intersections. We utilize several types of time-series data such as latitude, longitude, velocity, acceleration, rotation etc. This data is then processed to be used as input to deep learning models that can predict the future trajectory of a pedestrian according to his/her past trajectory. Deep learning is selected because it allows for predicting future pedestrian trajectory without making modeling assumptions. The collected trajectories are pre-processed into smaller trajectories that are uniform in size and converted into local geo-coordinates. To predict the future trajectory, a reachable set for each sub-trajectory is defined. Next, a framework composed of multiple deep-learning models is developed, the output of which is the trajectory of a pedestrian in the next few seconds.
The ultimate goal of this application is to alert people or vehicles of imminent dangerous situations and possibly recommend actions on how to avoid dangerous scenarios altogether, or reduce the severity of an upcoming incident if it cannot be avoided.
Google Slide Presentation
Presentation Script
Research Disciplines
Engineering