2018_BAA_Hazardous_Crowd_Dynamics__SOO_G.pdf

Hazardous Crowd Dynamics Detection Statement of Objectives BAA 18-02/Call 0001/SOO G

April 1, 2019 Science and Technology Directorate (S&T) Department of Homeland Security (DHS)

BAA 18-02/Call 0001/ SOO B Hazardous Crowd Dynamics Detection

Table of Contents

Contents

1.0 Problem Statement

2.0 Purpose and Scope

3.0 Background Environment

3.1 Mission Scenarios

4.0 Potential Solution Attributes

4.1 Concept of Operations (ConOps)

5.0 Solution Capabilities and Requirements

5.1 Must Have Requirements (Key Performance Parameters)

5.2 Desirable Capabilities or Characteristics

6.0 System Support

6.1 Contractual Requirements

6.1.1 Period and Place of Performance

6.1.2 System Affordability

6.1.3 Regulatory/Legal Requirements

6.1.4 Contract Deliverables

6.2 Solution Support

6.2.1 Training

6.2.2 Storage and Maintenance

1.0 Problem Statement

First responders lack the integrated technology to track hazardous behavior of individuals/groups during public events. First responder subject matter experts identified a need to detect hazardous behavior by people and crowds in public locations and during major events. Currently, personnel must manually review video streams and other visual data, if available, and typically are not doing so continuously throughout events.

2.0 Purpose and Scope

This technology will detect adverse behavior displayed by individuals or groups of people (crowds), including unsafe crowd speeds, density pressures, directions and other high risk behaviors that are synonymous with hazardous situations. In addition, the solution can leverage existing technologies to model safe crowd densities and likely crowd speeds overlapped from density and motion estimates from video analytics. It would need to dynamically account for barricades and security measures.

3.0 Background Environment

3.1 Mission Scenarios

This solution will directly affect the way first responders respond to hazardous behavior events of various crowd sizes. This will give first responders an opportunity to discern the root cause of the threat, identify where it is occurring and what level of response is needed to mitigate the issue.

Through early recognition and alerting via an app or other device, a response could be initiated in time to prevent/minimize escalation, have a more appropriate response (vs. generic call type or “just-in-case” dispatch), and potentially minimize impact. This, in turn, will save first responders valuable time, money and resources that could be allocated to respond to other situations.

Major events (sporting events, concerts, festivals, and more), as well as the crowds in attendance, would be much more secure with automated monitoring.

4.0 Potential Solution Attributes

This tool will be used to detect adverse behavior by individuals and crowds by recognizing changes in crowd speeds and density pressures including crushing, sudden running, an increase in packing or funneling into constrained areas, overly dense hot spots that need to be dispersed, and recognition of obstacles blocking flow. This is envisioned as an add-on to existing security infrastructure (video cameras from all sources, servers), which will reduce costs.

4.1 Concept of Operations (ConOps)

This solution strongly benefits the general public by allowing earlier threat identification; including scenarios where crowd behavior is not malicious, but just a threat indicator. It also benefits event organizers and first response agencies by reducing their safety risks. This is envisioned as an add-on to existing security infrastructure (video cameras from all sources, servers), which will help reduce cost.

Research and development efforts will likely take a phased approach, with the first-phase providing basic capabilities and subsequent phases adding in more sophisticated machine learning to detect more complicated situations. The projected phases for the project are as follows:

• Phase I: Develop a mechanism to assess crowd density—human analysts will look at the video feeds plus crowd density indicators to recognize potential problems. This phase also will entail assessing for behaviors (e.g., evacuation compliance) where the desired outcome is known.

• Phase II: Recognize anomalous individuals coming into a crowd; recognize other anomalies in crowd behavior that require immediate attention.

• Phase III: Recognize precursor behavior by individuals and crowd groups (similar to demonstrated flashover risk detection) to predict dangers that may happen within seconds or minutes.

5.0 Solution Capabilities and Requirements

5.1 Must Have Requirements (Key Performance Parameters)

The proposed solution must:

• Recognize sudden changes, from typical milling in crowds to the potential for unsafe pressures and directions. An example would be a change from random/typical behavior to anomalous behavior (e.g. running at once, conflicts, sudden changes of densities).

• Recognize agitators entering crowds that will likely pose a risk to crowd behavior, anomalous noise, and potential terror threats.

• Work from multiple angles and viewsheds (geographical areas visible from a specific location), including overhead views from buildings and UAVs.

• Help identify hot spots that need to be dispersed.

5.2 Desirable Capabilities or Characteristics

The proposed solution should:

• Leverage existing technologies to model safe crowd densities and likely crowd speeds overlapped from density and motion estimates from video analytics, taking into account barricades and security measures.

• Remotely monitor standard crowds (e.g. concert-goers, commuters in transit stations and on trains, crowded city squares and parks) as well as crowds at major events (e.g. protests, marches).

• Integrate with other safety alert systems (e.g. pressure sensor alerts, gunshot detection alerts) into agencies’ situational awareness tools.

6.0 System Support

6.1 Contractual Requirements

6.1.1 Period and Place of Performance

The length of time of the contract shall be between 15-18 months, and shall not exceed 18 months.

All developmental activities will be performed at the contractor-designated site(s). DHS may require the contractor to perform critical design reviews and/or product presentations at DHS-specified locations.

6.1.2 System Affordability

The solution will be cost-effective so first response agencies from a wide variety of jurisdictions will be able to afford it. First responder agency budgets are limited and should be taken into consideration when developing and determining a price-per-unit cost model.

6.1.3 Regulatory/Legal Requirements

DHS is committed to providing the first response community with information regarding technology availability. As a final deliverable, project performers will be required to submit applications for the technology solution to be included on the Federal Emergency Management Agency (FEMA) Authorized Equipment List and the Inter-Agency Board for Equipment Standardization and Interoperability Standardized Equipment List.

6.1.4 Contract Deliverables

DHS is committed to ensuring the success of this technology solution. Project performers will be required to provide DHS with eight prototype deliverables, which will be used for post-contract testing and evaluation.

DHS is also committed to ensuring first responder use of this technology. Project performers will also be required to provide DHS with documented metrics regarding product sales and estimated production costs for the first three years after commercialization. Additionally, to promote first responder technology awareness, DHS may request that the project performer participate in operational exercises, demonstrations, and conferences for the first three years after commercialization.

6.2 Solution Support

6.2.1 Training

Training information, as required, will be provided on any specialized equipment and will include how to properly store, clean, deploy, and maintain the basic ensemble/unit. Any specialty requirements of personnel conducting maintenance, storage, extraction, or delivery—including certification requirements—will be identified. Training material will be made available initially in hard-copy and subsequently in soft-copy format.

6.2.2 Storage and Maintenance

The contractor shall identify all necessary storage and maintenance requirements for the proposed solution including: the required maintenance schedule, identification of the specific components of the ensemble that requires maintenance, the type of maintenance required, and whether the maintenance can be performed in the field by operational personnel or if it will be required to be returned to the manufacturer for the required maintenance.

If the proposed solution is virtual, describe any storage or maintenance requirements, including archival activities, and the personnel performance needs associated with the solution.