Abstract
Using data collected and motion modelling developed in a previous project in this UTC, "Understanding and Guiding Pedestrian and Crowd Motion", as well as additional data to be collected, we will:
1. Develop pedestrian intention and motion modelling and prediction, with experimental validation,
2. Refine the sensor package and data analysis techniques of our pedestrian motion data collection system and datasets,
3. Develop vehicle’s motion planning and control algorithm for navigating, dodging, or stopping in pedestrian interaction scenarios.
Description
In a previous Project in this UTC Program, "Understanding and Guiding Pedestrian and Crowd Motion" we have both developed a model and simulation software for pedestrian and crowd motion and individual vehicles slowly moving within crowds. We have collected data and established a data base.
In this study, we shall focus on the effects on motion planning for automated vehicles interacting with pedestrians, as they step off the curb both individually and as groups.
In Year 1, we shall develop path/motion planning algorithms for vehicles autonomously maneuvering within crowds and dodging individual pedestrians. We shall investigate estimation techniques to find the probability and possibly future motion of static or moving pedestrians who might step off the curb (primarily at locations where there are no crosswalks). We shall develop both the hardware and software for a sensor package to use pedestrian detection from vehicles and do preliminary tests.
In year 2, we shall utilize the system developed in Year 1 to collect data and establish the effect of pedestrians on individual vehicles dodging or stopping and the implications on traffic flow. We shall refine the estimation, possibly using deep learning techniques or probabilistic approaches or combined, to build a full model for pedestrian motion prediction. We shall refine the path/motion planning algorithms to generate more efficient and safer vehicle maneuvers for dodging or navigating through individual pedestrians or crowds.
Timeline
In Year 1, we shall develop path/motion planning algorithms for vehicles autonomously maneuvering within crowds and dodging individual pedestrians. We shall investigate estimation techniques to find the probability and possibly future motion of static or moving pedestrians who might step off the curb (primarily at locations where there are no crosswalks). We shall develop both the hardware and software for a sensor package to use pedestrian detection from vehicles and do preliminary tests.
In year 2, we shall utilize the system developed in Year 1 to collect data and establish the effect of pedestrians on individual vehicles dodging or stopping and the implications on traffic flow. We shall refine the estimation, possibly using deep learning techniques or probabilistic approaches or combined, to build a full model for pedestrian motion prediction. We shall refine the path/motion planning algorithms to generate more efficient and safer vehicle maneuvers for dodging or navigating through individual pedestrians or crowds.
Strategic Description / RD&T
Deployment Plan
Expected Outcomes/Impacts
Expected Outputs
TRID
Individuals Involved
| Email |
Name |
Affiliation |
Role |
Position |
| ozguner.1@osu.edu |
Ozguner, Umit |
The Ohio State University |
PI |
Faculty - Tenured |
| redmill.1@osu.edu |
Redmill, Keith |
The Ohio State University |
Co-PI |
Faculty - Research/Systems |
Budget
Amount of UTC Funds Awarded
$133281.00
Total Project Budget (from all funding sources)
$250540.00
Documents
| Type |
Name |
Uploaded |
| Data Management Plan |
dmp-Ozguner-2020_GcY7Y0D.docx |
Jan. 6, 2020, 2:29 p.m. |
| Publication |
mpc_social_force-DSPIVS2018.pdf |
May 1, 2020, 10:44 a.m. |
| Presentation |
AVS19-poster.pdf |
May 1, 2020, 10:48 a.m. |
| Publication |
A Multi-State Social Force Based Framework for Vehicle-Pedestrian Interaction in Uncontrolled Pedestrian Crossing Scenarios |
Sept. 30, 2020, 6:40 p.m. |
| Progress Report |
317_Progress_Report_2020-09-30 |
Sept. 30, 2020, 6:42 p.m. |
| Publication |
Crowd motion detection and prediction for transportation efficiency in shared spaces. |
Dec. 2, 2020, 10:37 a.m. |
| Publication |
Top-view trajectories: A pedestrian dataset of vehicle-crowd interaction from controlled experiments and crowded campus. |
Dec. 2, 2020, 10:38 a.m. |
| Publication |
A Social Force Based Pedestrian Motion Model Considering Multi-Pedestrian Interaction with a Vehicle. |
Dec. 2, 2020, 10:41 a.m. |
| Publication |
A Multi-State Social Force Based Framework for Vehicle-Pedestrian Interaction in Uncontrolled Pedestrian Crossing Scenarios. |
Dec. 2, 2020, 10:42 a.m. |
| Publication |
Agent-based microscopic pedestrian interaction with intelligent vehicles in shared space. |
Dec. 27, 2020, 10:58 p.m. |
| Publication |
On the Generalizability of Motion Models for Road Users in Heterogeneous Shared Traffic Spaces |
April 13, 2021, 4:34 p.m. |
| Publication |
Predicting Pedestrian Crossing Intention with Feature Fusion and Spatio-Temporal Attention. |
March 27, 2022, 2:30 p.m. |
| Progress Report |
317_Progress_Report_2021-03-31 |
April 13, 2021, 4:40 p.m. |
| Publication |
Pedestrian Emergence Estimation and Occlusion-Aware Risk Assessment for Urban Autonomous Driving |
Oct. 6, 2021, 5:09 p.m. |
| Presentation |
Decentralized Control Problems in ITS |
Oct. 6, 2021, 5:18 p.m. |
| Progress Report |
317_Progress_Report_2021-09-30 |
Oct. 6, 2021, 5:18 p.m. |
| Publication |
Faraway-Frustum: Dealing with Lidar Sparsity for 3D Object Detection using Fusion |
March 27, 2022, 2:30 p.m. |
| Progress Report |
317_Progress_Report_2022-03-30 |
March 27, 2022, 2:31 p.m. |
| Publication |
Predicting Pedestrian Crossing Intention With Feature Fusion and Spatio-Temporal Attention |
Sept. 24, 2022, 10:29 p.m. |
| Publication |
On the Generalizability of Motion Models for Road Users in Heterogeneous Shared Traffic Spaces |
Sept. 24, 2022, 10:29 p.m. |
| Progress Report |
317_Progress_Report_2022-09-30 |
Sept. 24, 2022, 10:30 p.m. |
| Progress Report |
317_Progress_Report_2023-03-31 |
April 7, 2023, 8:45 a.m. |
| Final Report |
Final_Report_-_317.pdf |
Aug. 14, 2023, 7:16 a.m. |
Match Sources
No match sources!
Partners
| Name |
Type |
| Technische Universität Clausthal |
Deployment Partner Deployment Partner |
