Abstract
Vulnerable road users are considered people that are not in a vehicle and are, consequently, at a higher risk for serious injury because they have less crash protection than a vehicle occupant. Pedestrians, bicyclists, motorcyclists, and road workers are common vulnerable road users. Vulnerable road users can be further categorized by their degree of mobility, perception, and cognition. Shown in Figure 1 are numerous examples for each of the vulnerability categories. Vulnerabilities also have spatial and temporal dependencies, which can be quantified by a vulnerability index that might range from very low risk to very high risk (Figure 2). For example, running on a sidewalk in the middle of the day has an associated very low index. However, jaywalking across the road during rush hour might have a high or very high vulnerability index.
The goal of the proposed work is to enhance the safety of the vulnerable at intersections because these are locations of planned conflict and thus have an inherent risk. To accomplish this goal, we propose a cyber-physical system (Figure 3) that detects vulnerable road users, calculates a vulnerability index, then takes an appropriate action or actions to minimize the opportunity for injury. For example, if a person falls out of their wheelchair in the middle of a signalized intersection, all of the traffic signals would stay red, emergency medical vehicles would be dispatched, and audiovisual warnings would be broadcast. Warnings could also be sent via wireless communication to personal devices and even connected autonomous vehicles.
The core of the system is based on detecting the vulnerable in visual data captured from cameras. Accomplishing this task requires an annotated dataset of people with vulnerabilities, e.g., walking cane, bicycle, etc. There are some public datasets available with annotated wheelchair users for example, but not nearly enough vulnerable road users are available. To fill the dataset gap, we will deploy cameras in areas with expected high vulnerable activity. If we are unable to capture enough image examples, we will augment the dataset with synthetic images (e.g., project a 3D model of a man using crutches into an image) and/or perform enactments. Then we will train models for detecting the vulnerable, develop a method for calculating a vulnerability index, and develop a warning system.
We have a longstanding deployment partnership with the City of Pittsburgh Department of Mobility and Infrastructure (DOMI). DOMI has already approved a camera deployment at the intersection of Forbes and Morewood. We also have a relationship with Easterseals Massachusetts in an advisory capacity for issues related to people with vulnerabilities.
Description
Timeline
Strategic Description / RD&T
This project directly addresses the Data-Driven System Safety research priority of the RD&T Plan as specified below.
Safety Grand Challenge, Safe Design (page 19):
- Evaluate the safety performance of infrastructure design and develop and promote
the use of effective safety countermeasures
- Identify and support strategies to increase vulnerable road user safety (e.g., pedestrians,
bicyclists, motorcyclists, and people with disabilities).
Safety Grand Challenge, Safe Data (page 19): Research and develop new methodologies and tools for safety data collection, management, analysis, and evaluation:
- Develop safety data collection methods and advanced safety data and risk analysis techniques to identify and analyze emerging safety issues.
- Provide the scientific and engineering basis for policy decisions, improved industry standards, and enforcement and compliance matters.
- Assess safety incident trends and causes to enhance safety requirements and best practices.
Deployment Plan
Q1: Install cameras at Fifth and Morewood intersection (already approved by DOMI)
Q2: Capture images of intersection activity. Perform enactments as needed.
Q3-Q4: Assess whether images capture enough vulnerable users. If not, begin developing methods to create synthetic vulnerable users.
Expected Outcomes/Impacts
The result of this study is a system that identified vulnerabilities at an intersection then communicates that information to everyone at the intersection. This is achieved through wireless communication to personal devices or connected vehicles, audible alerts, and visual warnings. Widespread adoption of such a system would make navigating an intersection much safer for the vulnerable.
Expected Outputs
We anticipate the following outputs:
- Novel dataset of vulnerable road users
- Algorithms for detecting vulnerable road users
- Method for assigning a vulnerability index
TRID
Search keywords used were: detecting vulnerable road users intersection
A total of four records were returned.
- Paper 1: The appearance of vulnerable road users is altered. Assessment of how different parameters affect the detection of vulnerable road users. No work with intersections.
- Paper 2: Evaluation of how different non-intrusive detection technologies detect vehicles at signalized intersections in Minnesota. No work with the vulnerable.
- Paper 3: Developed a method to detect and measure evasive actions that pedestrians take during traffic conflicts. While a pedestrian is considered a vulnerable road user, they are easily detectable. Measuring evasive actions is not part of our proposed study.
- Paper 4: Developed a system for detecting and tracking objects and classifying them as either pedestrian, cyclist, or motor vehicle. Our proposed work will require a similar system. However, this paper was published in 2014 and thus the method is probably outdated.
Individuals Involved
| Email |
Name |
Affiliation |
Role |
Position |
| srinivas@cmu.edu |
Narasimhan, Srinivasa |
Carnegie Mellon Universit |
PI |
Faculty - Tenured |
| rtamburo@cmu.edu |
Tamburo, Robert |
Carnegie Mellon University |
Co-PI |
Faculty - Research/Systems |
| kvuong@andrew.cmu.edu |
Vuong, Khiem |
Carnegie Mellon University |
Other |
Student - PhD |
Budget
Amount of UTC Funds Awarded
$97194.00
Total Project Budget (from all funding sources)
$97194.00
Documents
Match Sources
No match sources!
Partners
| Name |
Type |
| Pittsburgh Department of Mobility and Infrastructure |
Deployment Partner Deployment Partner |
| Easterseals Massachusetts |
Equity Partner Equity Partner |
| Blind & Vision Rehabilitation Services of Pittsburgh |
Equity Partner Equity Partner |
| PathVu |
Deployment & Equity Partner Deployment & Equity Partner |
