Optimal Active Target Localisation Strategy with Range-only
Measurements
Shaoming He
a
, Hyo-Sang Shin
b
and Antonios Tsourdos
c
School of Aerospace, Transport and Manufacturing, Cranfield University, College Road, Cranfield, MK43 0AL, U.K.
Keywords: Target Localisation, Range-only Measurement, Optimal Manoeuvre.
Abstract:
This paper investigates the problem of one-step ahead optimal active sensing strategy to minimise estimation
errors with range-only measurements for non-manoeuvring target. The determinant of Fisher Information
Matrix (FIM) is utilized as the objective function in the proposed optimisation problem since it quantifies the
volume of uncertainty ellipsoid of any efficient estimator. In consideration of physical velocity and turning
rate constraints, the optimal heading angle command that maximises the cost function is derived analytically.
Simulations are conducted to validate the analytical findings.
1 INTRODUCTION
The past few years have witnessed an increasing inter-
est in the development and employment of small-scale
unmanned robots in both civil and military applica-
tions. One particular interesting mission of small-
scale unmanned robots is to track and localize targets
of interest in an automated fashion since reliable tar-
get tracking is a fundamental and key enabling tech-
nology in many practical applications, such as vehicle
navigation, situational awareness and public surveil-
lance (Atanasov et al., 2014; Salaris et al., 2017).
One main challenge for the operations of these small-
scale robots in target localisation is that they are typi-
cally constrained by limited payload, power and en-
durance. Therefore, only limited information, e.g.,
bearing-only or range-only, can be gathered due to the
limits of sensor availability.
It is well-known that the relative geometry be-
tween the observer and the target poses great effects
on the achievable localisation performance (Bishop
et al., 2010). For this reason, active sensing that finds
the optimal path or trajectory for information gain
maximisation can yield significant benefits to improv-
ing the perceptual results in target localisation (Bajcsy
et al., 2018). Through numerically maximising the
determinant of FIM over a finite horizon, optimal ob-
server trajectory for bearing measurements gathering
a
https://orcid.org/0000-0001-6432-5187
b
https://orcid.org/0000-0001-9938-0370
c
https://orcid.org/0000-0002-3966-7633
to localize a single target was proposed in Tokekar
et al. (2011). The rationale of leveraging FIM lies
in that it prescribes a lower bound of target localisa-
tion error covariance of any efficient estimation algo-
rithm (Taylor, 1979). Therefore, the determinant of
FIM can be utilized as a performance metric to quan-
tify the volume of the error uncertainty ellipsoid. Ex-
cept for FIM, the trace of error covariance, which di-
rectly quantifies the average estimation performance,
was also utilized to find the proper path for target
localisation using bearing-only measurements (Logo-
thetis et al., 1997; Zhou et al., 2011). Note that most
approaches utilize numerical methods to find the op-
timal solutions and consequently require high com-
putational burden. To alleviate the computational is-
sue, an analytical solution was proposed in (He et al.,
2019) using geometric analysis for bearing-only tar-
get localisation.
Apart from bearing-only-based target localisation,
range-only-based active sensing strategy is another
emerging low-cost solution for target localisation us-
ing small-scale robots due to the proliferation of
lightweight and low-cost LIDAR/infrared range find-
ers. Mart
´
ıNez and Bullo (2006); Bishop et al. (2010)
analysed the optimal relative target-observer geomet-
ric, that maximises system observability, for multiple
static sensors to localize a stationary target, where the
determinant of FIM was utilized as the cost function.
A range-only-based sliding mode controller was pro-
posed in Matveev et al. (2011) to drive a wheeled mo-
bile robot to a predefined distance from a manoeu-
vring target and makes the robot follow the target at
He, S., Shin, H. and Tsourdos, A.
Optimal Active Target Localisation Strategy with Range-only Measurements.
DOI: 10.5220/0007928400910099
In Proceedings of the 16th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2019), pages 91-99
ISBN: 978-989-758-380-3
Copyright
c
2019 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
91