Distributed collaborative localization for a heterogeneous multi-robot system

Wanasinghe, Thumeera R.; Mann, George K. I.; Gosine, Raymond G.

doi:10.1109/ccece.2014.6900998

Cited by 9 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Let (5) represent the set of child robots pose measurement densities that are generated by a given leader robot at time ; where represents the pose measurement density of a child robot as measured by a leader robot. Operator is the pose composition operator [5], [8], [40]. This pose composition is analogue to Cartesian-to-Cartesian coordinate conversion.…”

Section: Child Robots' Pose Measurementmentioning

confidence: 99%

“…where (8) represents the set of relative pose measurement densities that corresponds to the relative pose measurements acquired by all the members of the MRS at time step . The leader can then construct the RPMG in the global perspective.…”

Section: Mrs With a Single Leader Robot And Multiple Child Robotsmentioning

confidence: 99%

“…If the measurements satisfy the measurement gating condition, the leader robot fuses these measurements with its current state estimation in order to improve its localization (line 7). Otherwise, the predictive density is directly assigned to the posterior density of the state estimation (line [8][9]. In this way, we can identify and dismiss any outliers.…”

Section: A Leader Robots Localizationmentioning

confidence: 99%

“…The objective of a leader-assistive localization scheme is establishing the localization for the child robots using the leader robot pose and inter-robot observations between the leader robots and the child robots [7]. Child robots can be localized either on a given reference coordinate frame [8], or on the leader robot's body-fixed coordinate frame [6]. In order to execute such a leader-assistive navigation framework, child robots should always navigate within the sensing range of the leader robots.…”

Section: Introductionmentioning

confidence: 99%

“…Child robot's pose can be estimated either on a given coordinate frame (or global coordinate frame) [8] or on the body-fixed coordinate system of leader robots [1], [29], [30]. The former can either be performed in a distributed manner or centralized manner.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Distributed Leader-Assistive Localization Method for a Heterogeneous Multirobotic System

Wanasinghe

Mann

Gosine

2015

IEEE Trans. Automat. Sci. Eng.

Self Cite

View full text Add to dashboard Cite

This paper presents a distributed leader-assistive localization approach for a heterogeneous multirobotic system (MRS). The localization algorithm is formulated to estimate the position and orientation (pose) of a group of robots in a given reference coordinate frame (or global coordinate frame). It is assumed that the heterogeneous-MRS has one or a group of robots (which we refer as leader robots) with higher sensor payload, higher processing power, and larger memory capacity, enabling accurate self-localization capabilities. Robots with limited resources (which we refer as child robots) rely on leader robots, and inter-robot observations between leaders and themselves for localization. Finite-range sensing is a key challenge for such leader-assistive localization. This study presents a sensor sharing technique which virtually enhances the sensing range of leader robots. In the proposed method, each robot locally runs a cubature Kalman filter to estimate its own pose and hosts a low cost, lightweight, and low-power sensory system to periodically measure relative pose of neighbors. Each robot transmits these relative pose measurements to leader robots. Leader robots then combine available relative observations in order to synthesise global pose measurements and associated noise covariances for child robots. Child robots are acknowledged by the leader robots with the synthesized global pose measurements and fuse these measurements with their local belief in order to improve their localization. Theoretical developments are presented to virtually enhance the leader robots' sensing range. The performance of the proposed distributed leader-assistive localization algorithm is evaluated on a multirobot simulation test-bed and on a publicly available multirobot localization and mapping data-set. The results illustrate that the proposed algorithm is capable of establishing accurate and consistent localization for the child robots even when they operate beyond the sensing range of the leader robots.Note to Practitioners-MRS can be used to perform environmental monitoring, exploration tasks, and search and rescue missions. Accurate localization is a critical factor that governs the success of the autonomous mobile robots-based missions. In order to improve the localization accuracy, robots can be equipped with advanced sensory systems which will increase the cost. Additionally, robots can execute advanced localization algorithms to generate an accurate localization which entails higher processing capabili-

show abstract

Section: Child Robots' Pose Measurementmentioning

confidence: 99%

Section: Mrs With a Single Leader Robot And Multiple Child Robotsmentioning

confidence: 99%

Section: A Leader Robots Localizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%