For decades, the effects of climate change have melted Polar ice caps, driving polar bears to spend more time on land near Canada’s northernmost communities. As a result, bears and humans have been interacting more frequently—encounters that can be dangerous or even devastating for both parties. A new technology referred to as “bear-dar” is being tested to determine if it can help humans avoid contact with bears. Bear-dar uses a radar system that detects when polar bears are headed towards towns, and sends out an early warning signal for people nearby.
So far, bear-dar has shown great potential in helping to prevent run-ins between humans and polar bears. The technology being used is evolving to improve accuracy and address challenges – for example, bear-dar units may detect motion from another large animal, like a caribou, and send out the warning signal needlessly. Artificial intelligence is now being incorporated to help the units detect whether the motion is that of a polar bear or an animal that poses no danger to humans.
The standards community supports many of the technologies used in the bear-dar units. IEEE 686-2017, IEEE Standard for Radar Definitions, is an American National Standard (ANS) developed by IEEE that promotes clarity and consistency in the use of radar terminology, representing the consensus of a panel of radar experts. The artificial intelligence now being used to fine-tune the devices is guided by standards as well, such as ANSI/CTA 2096-2021, Guidelines for Developing Trustworthy Artificial Intelligence Systems. Developed by the Consumer Technology Association (CTA), this ANS describes the principles of trustworthy artificial intelligence – an area of great concern when AI is used to protect human and animal lives, as it is in this application. IEEE and CTA are members and accredited standards developers of the American National Standards Institute (ANSI).
As impressive as they are, these early model units are very expensive to produce, making them out of reach for some communities. A smaller, cheaper device is also being developed that relies on lidar rather than radar. The lidar device has a shorter range, but is more portable and could bring a similar bear-detecting system to communities with fewer financial resources. INCITS/ISO TS 19130-2:2014(R2020), Geographic Information – Imagery Sensor Models for Geopositioning – Part 2: SAR, InSAR, Lidar and SonaR, is a technical specification prepared by INCITS that specifies the sensor models and metadata for geopositioning images remotely sensed by lidar, among other systems. INCITS is an ANSI member and accredited standards developer.
Ultimately, the bear-dar radar and lidar systems may not become the singular solution to prevent bear-human encounters; it’s more likely that they will be used in tandem with other safety procedures. Regardless, they have the potential to become an important part of the systems that protect humans and polar bears in these northern habitats. Read more in Hakai Magazine and in an NPR “Weekend Edition Sunday” segment.