UST 031

92 W hile many UGVs are developed to do specific jobs for the military or industry (or both), some have a mission to advance robotics itself while helping to educate the next generation of unmanned vehicle developers in universities and other organisations focused on research, development and learning. This is what Clearpath Robotics created the Jackal for. A stablemate to the company’s larger Husky UGV, the Jackal is a portable robotic research platform that will fit easily into the back of a modestly sized car. Light enough, at 17 kg, to be carried by one person, it is essentially a battery- powered box on four wheels that is rugged enough for outdoor use yet small enough for indoor use and which offers sufficient power and payload capacity for practical research work. To Clearpath, that meant being able to deploy full x86 computing hardware, including graphics processing units, as well as supporting the Robot Operating System (ROS). It also allows a range of sensing and actuation hardware to be integrated, explains the company’s engineering research solutions manager Robbie Edwards. During 2013-14, he says, Clearpath’s customers and the wider robotics community were beginning to express an interest in outdoor multi-robot operations, so the company partnered with the US Army Research Laboratory to develop a low-cost platform that could work alone or as part of a swarm. “Requirements for the platform included being water-resistant to the point that our users don’t have to worry about driving through puddles, mud, snow or being stuck in light rain,” Edwards says. Other key requirements were to have the space and power to support a mini- ITX format computer (based on a 17 x 17 cm motherboard) and to integrate Lidars such as the Sick LMS-111 and the Velodyne VLP-16, along with a range of cameras and networking equipment. “The Jackal opens up so that this Peter Donaldson explains how this UGV was developed to meet the needs of robotics research groups Research facility April/May 2020 | Unmanned Systems Technology