Estimated emergency vehicle accidents in the United States cost $35 billion annually, and fatalities caused by collisions are 4.8 times higher for emergency responders than the national average. Police officers have roughly double the rate of motor vehicle crashes per million vehicles driven than the public.
During a session at IACP 2021, panelists defined the problems and discussed potential solutions being worked on by Thomas Lu, Ph.D., and Edward Chow, Ph.D., of NASA’s Jet Propulsion Lab (JPL) at the California Institute of Technology.
Trusted and Explainable Police Artificial Intelligence
The sheer volume of data from next-generation communication tools and sensors risks overwhelming or distracting first responders from critical activities. Information overload can create obstacles for first responders to perform their duties safely and efficiently. This situation applies not only to first responders on the ground but also to those tasked with managing and directing an incident response from a higher level.
Identifying this as an issue, JPL – which has been developing artificial intelligence (AI) platforms for use during space exploration – has been experimenting with transitioning these into the public safety space for the last three to four years. JPL contends that its developed AI products could enhance officer safety through enriched 360-degree situational awareness.
AI assistant
To provide solutions, JPL is conducting a project funded by the US Department of Transportation National Highway Traffic Safety Administration to study technology for improving the safety of first responder and roadside crews in and around active traffic. This resulted in the creation of the Trusted and Explainable Police Artificial Intelligence (TruePAL) AI assistant, which provides real-time warnings of risks by analyzing the environment and traffic patterns to generate a timely warning to drivers and roadside crews to avoid crashes.
In simple terms, TruePAL is being developed to be an AI system that examines and manages all inputs from onboard distractions such as radio, radar computers, in-car sensors, scanners, cameras, GPS navigation and provides 360-degree situational awareness and recommendations for action. This will all be fed back to the driver via a Heads-Up Display (HUD) and a voice interface. In response, the driver will be able to communicate back with the onboard systems via voice commands using a chatbot (think Alexia or Siri!).
The TruePAL team comprised members from NASA’s JPL, Temple University and the Miami Dade (Florida) Police Department and was conducted in two phases. First, stakeholders were engaged to identify key challenges and use cases and TruePAL was developed and tested in a simulated environment to demonstrate feasibility, followed by testing and validation of TruePAL with real first responder traffic crash data.
Computer scenario testing
To develop AI responses within the TruePAL system, JPL developed computer models using the CARLA driving simulator of several common scenarios that may place vehicles and crews at risk. These were:
- Intersection safety: Involving police, EMS, fire and wrecker agencies on the way to an accident scene. TruePAL modeled assisting drivers to cross the intersections.
- Roadside safety: TruePAL provides 360-degree situational awareness for roadside parked vehicles and warns vehicles of a potential collision.
- Hazard Sign ID: TruePAL detects the truck’s hazard sign and transmits information to the relevant actors about hazardous or flammable materials.
- First aid assistant: TruePAL guides responders to prioritize first aid and to perform corrective procedures.
- Electric vehicle guide: TruePAL guides the responder to correctly handle the electric vehicle and battery involved in the crash.
From the modeling of each scenario, improvements and enhancements can be made to the TruePal system.
The future of TruePAL
Once developed, JPL’s technology will provide a cutting-edge AI tool to make public safety agencies inherently safer. Sensors and systems, along with the connected alarms and alerts, will allow officers to conduct all the necessary tasks required of them while they keep their hands on the wheel and eyes on the road.
The JPL team continues its research and is seeking feedback and use cases to expand the capability of these AI systems. Contact Thomas.t.lu@jpl.nasa.gov and Edward.chow@jpl.nasa.gov.
