This paper investigates approaches to deliberately designing systems whose controllability can be... more This paper investigates approaches to deliberately designing systems whose controllability can be quantified. Preliminary findings of ongoing research are presented on complex dynamical system control algorithms. The specification analysis and quality of the pressure control algorithm applied to a Topical Negative Pressure Wound Therapy device are conducted, with further discussion on self-regulation mechanism and characterization of both the partially observable and partially controllable workspace represented by the negative pressure chamber. Statistical methods are employed to understand the device physics and fuzzy logic and bacterial memetic algorithm are utilised to explore and optimize the existing algorithms and also extract the rule base.
Fuzzy rule-based model for outlier detection in a Topical Negative Pressure Wound Therapy Device
ISA Transactions, 2021
This paper proposes a novel method for offline outlier detection in nonlinear dynamical systems u... more This paper proposes a novel method for offline outlier detection in nonlinear dynamical systems using an input-output dataset of a Topical Negative Pressure Wound Therapy Device, NPWT. The fundamental characteristics of an NPWT describe a chaotic system whose states vary over time and may result in unpredictable and possibly anomalous divergent behavior in the presence of perturbations and other unmodeled system dynamics, despite a quasi-stable controller. Bacterial Memetic Algorithm, BMA, is used to generate fuzzy rule-based models of the input-output dataset. The error definition in the fuzzy rule extraction features a novel application of the Canberra Distance. The optimal number of rules for identifying the outliers, validated against both artificial and real system datasets, is calculated from the sample of inferred fuzzy models. The optimal number of rules is two in both cases based on the maximum average-error-drop. Using three or more rules results in better error performance; however, the algorithm learns the nuances of the outlier patterns instead. Novel methods for creating the outlier list and determining the optimal number of rules for the outlier detection problem are proposed.
2016 International Symposium on Micro-NanoMechatronics and Human Science (MHS), 2016
In this paper we continue with previous work by the authors implementing context-aware middleware... more In this paper we continue with previous work by the authors implementing context-aware middleware to accelerate robot learning from demonstration, LfD. Specifically, we apply Fuzzy Q-Learning, FQL, reinforcement learning strategy to enhance the learning experience of the robot. Typically, fuzzy techniques allow the robot to make decisions without the need for an exhaustive map of the world. FQL, approximates the observable configuration space allowing the robot to overcome the high dimension challenge of feature decomposition and navigation in a stochastic environment.
In this paper we propose the creation of context-aware middleware to solve the challenge of integ... more In this paper we propose the creation of context-aware middleware to solve the challenge of integrating disparate incompatible systems involved in the teaching of human action skills to robots. Contextaware middleware provides the solution to retrofitting capabilities onto existing robots (agents) and bridges the technology differences between systems. The experimental results demonstrate a framework for handling situational and contextual data for robot Learning from Demonstration.
This paper investigates approaches to deliberately designing systems whose controllability can be... more This paper investigates approaches to deliberately designing systems whose controllability can be quantified. Preliminary findings of ongoing research are presented on complex dynamical system control algorithms. The specification analysis and quality of the pressure control algorithm applied to a Topical Negative Pressure Wound Therapy device are conducted, with further discussion on self-regulation mechanism and characterization of both the partially observable and partially controllable workspace represented by the negative pressure chamber. Statistical methods are employed to understand the device physics and fuzzy logic and bacterial memetic algorithm are utilised to explore and optimize the existing algorithms and also extract the rule base.
Fuzzy rule-based model for outlier detection in a Topical Negative Pressure Wound Therapy Device
ISA Transactions, 2021
This paper proposes a novel method for offline outlier detection in nonlinear dynamical systems u... more This paper proposes a novel method for offline outlier detection in nonlinear dynamical systems using an input-output dataset of a Topical Negative Pressure Wound Therapy Device, NPWT. The fundamental characteristics of an NPWT describe a chaotic system whose states vary over time and may result in unpredictable and possibly anomalous divergent behavior in the presence of perturbations and other unmodeled system dynamics, despite a quasi-stable controller. Bacterial Memetic Algorithm, BMA, is used to generate fuzzy rule-based models of the input-output dataset. The error definition in the fuzzy rule extraction features a novel application of the Canberra Distance. The optimal number of rules for identifying the outliers, validated against both artificial and real system datasets, is calculated from the sample of inferred fuzzy models. The optimal number of rules is two in both cases based on the maximum average-error-drop. Using three or more rules results in better error performance; however, the algorithm learns the nuances of the outlier patterns instead. Novel methods for creating the outlier list and determining the optimal number of rules for the outlier detection problem are proposed.
2016 International Symposium on Micro-NanoMechatronics and Human Science (MHS), 2016
In this paper we continue with previous work by the authors implementing context-aware middleware... more In this paper we continue with previous work by the authors implementing context-aware middleware to accelerate robot learning from demonstration, LfD. Specifically, we apply Fuzzy Q-Learning, FQL, reinforcement learning strategy to enhance the learning experience of the robot. Typically, fuzzy techniques allow the robot to make decisions without the need for an exhaustive map of the world. FQL, approximates the observable configuration space allowing the robot to overcome the high dimension challenge of feature decomposition and navigation in a stochastic environment.
In this paper we propose the creation of context-aware middleware to solve the challenge of integ... more In this paper we propose the creation of context-aware middleware to solve the challenge of integrating disparate incompatible systems involved in the teaching of human action skills to robots. Contextaware middleware provides the solution to retrofitting capabilities onto existing robots (agents) and bridges the technology differences between systems. The experimental results demonstrate a framework for handling situational and contextual data for robot Learning from Demonstration.
Uploads
Papers by Charles Phiri