A System for Tracking an Autonomously Controlled Canine

Abstract

This paper presents an approach for outdoor navigation of an autonomously guided canine using an embedded command module with vibration and tone generation capabilities and an embedded control suite comprised of a microprocessor, wireless radio, GPS receiver, and an Attitude and Heading Reference System. In order to determine the canine's motions, which inherently contain non-conventional noise characteristics, the sensor measurements were integrated using a specialized Extended Kalman Filter (EKF), equipped with a Fuzzy Logic controller for adaptive tuning of the Process Noise Covariance Matrix. This allowed for rejection of un-modelled canine motion characteristics which tend to corrupt accelerometer bias tracking in a standard EKF. The EKF solution provided an optimized estimate of the canine position and velocity and also proved to be effective in tracking the canine's position (within 7·5 m) and velocity (within 1·2 m/s) during simulated 10 second GPS outages. K E Y

The integration of dogs into collaborative human-robot teams: an applied ethological approach

“…They carried out and described the results of a few pilot trials on a single, trained SAR dog during repeated remote command trials. They report on differences in the sensor signals corresponding to three investigated dog activities (stationary, walking and running), while the emphasis of their work was rather on the most accurate position tracking (with GPS) (Britt et al, 2011;Miller et al, 2012). Brugarolas and colleagues (2013) aimed to differentiate between static postures (sitting, standing, lying down, standing on two legs and eating off the ground) and dynamic activities (walking, climbing stairs and walking down a ramp) of dogs (N=7, of which 6 Labrador retrievers and one Kai Ken) using tri-axial accelerometer and tri-axial gyroscope and computational modeling (machine learning algrorithm).…”

“…'stop', 'turn left', etc. (Miller et al, 2012;Rybarcyzk et al, 2015). If reliable and delicate enough, such a sytem can also be useful in training contexts, since the handler can receive immediate feedback about the dogs' obedience if out of sight, while monitoring the animals' welfare status too (Bozkurt et al, 2014).…”

“…The reason for chosing these was because they cover the very basic behavioural repertoire typical of a dog moving on open field (also SAR dogs, see e.g. Miller et al, 2012), and the nonoverlapping nature of the categories was an important aspect in order to provide more clearcut conditions for the first steps of the automatic behaviour recognition.…”

“…Our new generation logger device, on the other hand, contains a magnetometer as well, with which a full attitude and heading reference system (AHRS) (Chang and Lin, 2011) is already available. It had been shown that using magnetometers and accounting for pitch in the motion model improved the tracking solution of canines during brief GPS outages (Miller et al, 2012). Miller and colleagues (2012) note as well that 'accelerometers are notorious for significant long-term error due to their inherent biases, scale factors, and sensitivity to noise'.…”

A Neurodynamics Control Strategy for Real-Time Tracking Control of Autonomous Underwater Vehicles

“…The fuzzy control approach (Wai, 2007; Ishaque et al, 2010, 2011; Miller et al, 2012) is employed for controlling uncertain or strongly nonlinear systems without knowing the precise system model. The fuzzy control approach does not need on-line or off-line learning and the computation procedure is relatively simple.…”