Inside Unmanned Systems

DEC 2017 - JAN 2018

Inside Unmanned Systems provides actionable business intelligence to decision-makers and influencers operating within the global UAS community. Features include analysis of key technologies, policy/regulatory developments and new product design.

Issue link: http://insideunmanned.epubxp.com/i/917103

Contents of this Issue

Navigation

Page 33 of 67

34 unmanned systems inside   December 2017/January 2018 These detections and characteristics pass through what is called a target classifier, and then probabilities are set up in the form of percentages, for example, they might re- port 80 percent likely to be a UAS and 20 percent likely to be a bird. With this data alarms can be set based on a per- centage threshold being reached. For example, and alarm can can be set so that anything greater than 80 percent likely to be a drone, take action Using classifiers can give confidence so that you can use more informed prognoses to trigger alarms. The classifier is reinforced in realtime by by the operator and cross-validation and performance metrics generated within 60 seconds. "Performance improves over time with increased true posi- tive and true negative classification rate," Romero said. Cameras & Video Tracking Lamm described the role played by cameras and video in drone detection, and he too stressed the importance of using a layered approach. Cameras, however, do have shortcomings, according to Lamm. These include the fact that cameras are not good at initial detection while trying to cover a large area (that is, they are not good as wide area situational awareness, an area in which RF detection and radar excel). But there are clear strengths from camera use in CUAS: 1. Zoom in verify close to allow visual identif ication (drone vs. bird, dangerous payload or not), Often times v isual identif ication is necessar y to allow countermeasures. 2. Cameras along with video target tracking can point ef- fectors very accurately. Examples of effectors include jammers, spotlights, net guns, and kinetic weapons. "We can actually get eyes on targets and visually deter- mine," Lamm said. "Video tracking a drone, and following an object even though it's moving in different directions aggressively (can be achieved)." These methods provide plenty of information to assists decision-making in terms of taking action. EO (Electro Optical) Daylight Camera, LWIR (Long Wave Infrared) Uncooled Thermal, and MWIR (Mid Wave Infrared) Cooled Thermal offer different benefits. Since objects tend to "pop out" in thermal imagery against the sky background, Lamm demonstrated some of the benefits provided by thermal imagery. Mitigation was addressed brief ly in the webinar, but Poss reminded the attendees that policy in many coun- tries—certainly in the United States—prohibit mitigation in most cases. "Really the only people who can engage in counter UAS (in the States) is the DoD, secret service when life of president involved, the Department of Energy when nuclear facilities are involved…and there is still a ranging legal battle over who has the right to listen in and decode RF signals." WEBINAR RECAP » SECURITY RISKS Photo courtesy of Black Sage Technologies. Drone detection in action at a November demonstration with R&S Ardronis and Black Sage at a military operating area.

Articles in this issue

Archives of this issue

view archives of Inside Unmanned Systems - DEC 2017 - JAN 2018