Adaptive Behaviour for a Self-Organising Video Surveillance System Using a Genetic Algorithm

Fabrice Saffre,Hanno Hildmann

doi:10.3390/a14030074

Abstract

Genetic algorithms (GA’s) are mostly used as an offline optimisation method to discover a suitable solution to a complex problem prior to implementation. In this paper, we present a different application in which a GA is used to progressively adapt the collective performance of an ad hoc collection of devices that are being integrated post-deployment. Adaptive behaviour in the context of this article refers to two dynamic aspects of the problem: (a) the availability of individual devices as well as the objective functions for the performance of the entire population. We illustrate this concept in a video surveillance scenario in which already installed cameras are being retrofitted with networking capabilities to form a coherent closed-circuit television (CCTV) system. We show that this can be conceived as a multi-objective optimisation problem which can be solved at run-time, with the added benefit that solutions can be refined or modified in response to changing priorities or even unpredictable events such as faults. We present results of a detailed simulation study, the implications of which are being discussed from both a theoretical and practical viewpoint (trade-off between saving computational resources and surveillance coverage).

Highlights

The Internet of Things (IoT) frequently involves retrofitting existing infrastructure with “intelligent” capabilities, see, e.g., in [1]
Genetic Algorithms (GAs)—which are among the “most well-regarded evolutionary algorithms in history” [17] are mostly used as an offline optimisation method to discover a suitable solution to a complex problem prior to implementation
We investigated how our proposed genetic algorithm framework would deal with each of the three limit-cases, and for both reproduction methods

Summary

Introduction

The Internet of Things (IoT) frequently involves retrofitting existing infrastructure with “intelligent” capabilities, see, e.g., in [1] This requires identifying methods for turning an ad hoc collection of newly “connected” devices [2] into a cohesive whole capable of providing a better service than was previously achievable, due to the isolation of its various components [3]. We aim to optimise the system’s performance (i.e., the cost for—and the benefit generated from—operating it) by scheduling the individual camera’s activation. This is similar to solving a multi-objective optimisation problem:. We apply the GA paradigm to the run-time adaption [23] of schedules for an ad hoc collection of cameras

Schedule Encoding and Fitness Function

Limit-Cases and Benchmarking

Exploration of the Weighting Parameters Space

Run-Time Adaptation and Real-World Considerations

Findings

Conclusions and Future Work