An Event-Triggered Distributed Continuous-Time Optimization Approach

Min Hu,Qiang Lu,Xue-Na Qiu,Qiuxuan Wu

doi:10.1109/access.2019.2939643

Min Hu, Qiang Lu + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2939643

Copy DOI

Abstract

In this paper, we design an event-triggered distributed continuous-time optimization approach. The proposed approach can be implemented independently by each robot and has two main characteristics. One is that each robot can deal with its local cost function such that the minimum of the sum of all the local cost functions can be found. The other is that, due to using the event-triggered communication mechanism, resource consumption of chips are saved by reducing the communication frequencies and the updating times of control inputs before the consensus is arrived. Moreover, based on Lyapunov theory, the stability conditions of multi-robot systems with the proposed approach are given. Finally, the effectiveness of the proposed approach is illustrated through experimental results.

Highlights

In multi-robot systems, distributed cooperative control has attracted extensive research because of wide application [1]–[5], [8], [12], [14], [16], [23], [25]–[28], such as stochastic source seeking [1], [20], [29], environmental monitoring [3], [4], [13], and odor source localization [11], [12] and so on
Signal source localization as a typical application of multi-robot systems has been well studied based on the distributed cooperative control approaches [17]–[19], [24]
We have investigated the problem of signal source localization by analyzing and designing an event-triggered distributed continuous-time optimization approach for multi-robot systems

Summary

INTRODUCTION

In multi-robot systems, distributed cooperative control has attracted extensive research because of wide application [1]–[5], [8], [12], [14], [16], [23], [25]–[28], such as stochastic source seeking [1], [20], [29], environmental monitoring [3], [4], [13], and odor source localization [11], [12] and so on. Signal source localization as a typical application of multi-robot systems has been well studied based on the distributed cooperative control approaches [17]–[19], [24]. M. Hu et al.: Event-Triggered Distributed Continuous-Time Optimization Approach obtain states information. The purpose of this paper is to design an event-triggered distributed continuous-time optimization method based on multiple mobile robots and apply it to the problem of signal source localization. (1) A distributed continuous-time optimization method based on event-triggered communication is proposed, where it reduces the updating frequencies of controller and relieves the communication burden between robots. (2) The performance conditions for the multi-robot system with the proposed event-triggered distributed continuous-time optimization approach are presented by employing Lyapunov theory. (3) Through the physical experiment of signal source localization, the performance capability of this event-triggered distributed continuous-time optimization approach is illustrated. Wi L 2 where vil and vir represent the left and right wheels’ speed of robot i, respectively

PROBLEM FORMULATION

SIGNAL SOURCE LOCALIZATION

PERFORMANCE METRICS

CONCLUSION