Skip to main content

Ferdian Jovan

Followers

17

Following

3

Co-authors

2

Public Views

I am a Postdoctoral Researcher in Department of Computer Science, Royal Holloway University of London.My research focus is about utilising temporal planners for multi-robot coordination in operations and maintenance applications specifically offshore wind farms. I devised a reactive and adaptive planning system that underpins the execution of coordinated tasks, which are carried out by heterogeneous robots. I was a PDRA in Department of Engineering Science, University of Oxford where I created generative models for predicting battery health and its remaining useful life on real-world data. I was part of the STRANDS project where I modeled re-occurring patterns of human activity for autonomous mobile robots in an extended period of time.

less

Interests

Uploads

Papers by Ferdian Jovan

Adaptive Temporal Planning for Multi-Robot Systems in Operations and Maintenance of Offshore Wind Farms

Proceedings of the AAAI Conference on Artificial Intelligence

With the fast development of offshore wind farms as renewable energy sources, maintaining them ef... more With the fast development of offshore wind farms as renewable energy sources, maintaining them efficiently and safely becomes necessary. The high costs of operation and maintenance (O&M) are due to the length of turbine downtime and the logistics for human technician transfer. To reduce such costs, we propose a comprehensive multi-robot system that includes unmanned aerial vehicles (UAV), autonomous surface vessels (ASV), and inspection-and-repair robots (IRR). Our system, which is capable of co-managing the farms with human operators located onshore, brings down costs and significantly reduces the Health and Safety (H&S) risks of O&M by assisting human operators in performing dangerous tasks. In this paper, we focus on using AI temporal planning to coordinate the actions of the different autonomous robots that form the multi-robot system. We devise a new, adaptive planning approach that reduces failures and replanning by performing data-driven goal and domain refinement. Our experi...

Mongodb Store: Logging, Storing, Accessing Mongodb Through Ros

MongoDB tools for storing and analysing runs of ROS systems.

Multimodal Indoor Localisation for Measuring Mobility in Parkinson's Disease using Transformers

Cornell University - arXiv, May 12, 2022

Parkinson's disease (PD) is a slowly progressive debilitating neurodegenerative disease which is ... more Parkinson's disease (PD) is a slowly progressive debilitating neurodegenerative disease which is prominently characterised by motor symptoms (e.g. bradykinesia, rigidity and tremor). Indoor localisation, including number and speed of room to room transitions, provides a proxy outcome which represents mobility and could be used as a digital biomarker to quantify how mobility changes as this disease progresses. We use data collected from 10 people with Parkinson's, and 10 controls, each of whom lived for five days in a smart home with various sensors. In order to more effectively localise them indoors, we propose a transformerbased approach utilizing two data modalities, Received Signal Strength Indicator (RSSI) and accelerometer data from wearable devices, which provide complementary views of movement. Our approach makes asymmetric and dynamic correlations by a) learning temporal correlations at different scales and levels, and b) utilizing various gating mechanisms to select relevant features within modality and suppress unnecessary modalities. On a dataset with real patients, we demonstrate that our proposed method gives an average accuracy of 89.9%, outperforming competitors. We also show that our model is able to better predict in-home mobility for people with Parkinson's with an average offset of 1.13 seconds to ground truth.

A multirobot system for autonomous deployment and recovery of a blade crawler for operations and maintenance of offshore wind turbine blades

Journal of Field Robotics

Pengenalan Dan Pencarian Posisi Robot Dalam Pencarian Sumber Kebocoran Gas

Implementasi algoritma Particle Swarm Optimization (PSO) pada robot untuk pencarian sumber keboco... more Implementasi algoritma Particle Swarm Optimization (PSO) pada robot untuk pencarian sumber kebocoran gas memerlukan informasi posisi dari setiap robot. Penggunaan kamera memungkinkan untuk mendapatkan informasi posisi robot yang lebih absolut dan akurat dibandingkan dengan perangkat pencari informasi posisi lainnya. Paper ini memaparkan suatu metode yang akan digunakan untuk mencari posisi robot. Metode yang diajukan menggunakan beberapa teknik pengolahan citra, seperti, Color Filtering dan Blobs Filtering guna mengenali dan memperoleh posisi robot. Berdasarkan eksperimen yang telah dilakukan, metode ini mampu menentukan posisi absolut robot sehingga ia dapat menyediakan informasi posisi robot dalam implementasi algoritma PSO. Implementation of Particle Swarm Optimization (PSO) algorithm on the robot to search for the gas leakage source requires information of the position of each robot. The use of the camera allows the robot to obtain position information more accurate than the abs...

Real Multiple Mobile Robots Implementation of PSO Algorithm for Odor Source Localization

Many Researchs in robotics area to localize the odor source have been conducted recently. Simulat... more Many Researchs in robotics area to localize the odor source have been conducted recently. Simulation become the priority to be used instead of real hardware implementation. Despite many researchs are limited to the simulation result, this paper tries to verify the robustness of Particle Swarm Optimization (PSO) algo- rithm in the real-world implementation. This algorithm is implemented to localize a single odor source using the real hardware. A group of mobile robots equipped with wireless communication device and odor sensors is employed. A set of camera is used to track the position of each robots. The result verifies that PSO is technically sound for real-world odor source localization.

Penggunaan berbagai jenis gas dalam kehidupan sehari-hari turut andil dalam meningkatkan kenyamanan hidup manusia saat ini. Namun, peningkatan ini juga diikuti dengan meningkatnya bahaya yang muncul akibat ketidakhati-hatian dalam penggunaan gas, seperti kebocoran tabung atau pipa. Sebagai contoh, k

Odor source localization is a research which combines the ability to recognize and classify odor ... more Odor source localization is a research which combines the ability to recognize and classify odor fragrance with autonomus localization method. This paper will present some experiment results regarding odor source localization using a single agent or multi agents. Our experiment shows that single agent approach is not suitable to overcome odor localization in a dynamic environment. Hence, we propose a new method based on the Particle Swarm Optimization (PSO) and it’s modifications to solve the problem. Our research experiment shows that PSO success where single agent failed. The modifications aim to enhance PSO so that it can be used to localize multiple odor sources. All of the modifications are verified using our simulation software and/or Al-Fath robot. The result shows that our approach can localize four odor sources inside 7x7 m room in 13,216 minutes.

Robots implementation for odor source localization using PSO algorithm

WSEAS Transactions on Circuits and Systems archive, 2011

Research topics in robotic application is quite varies but one of the most interesting topic is o... more Research topics in robotic application is quite varies but one of the most interesting topic is odor source localization. This research combine the robot ability to recognize odor and track the movement so that robot can find the source. Most of the research are done to improve the algorithm to localize the source by using simulation software. This paper tries to verify the robustness of one of the localization method known as Particle Swarm Optimization (PSO) in the real-world implementation. This paper will shows robot model that used in the experiment and also discuss the architecture to implement robot behavior. A group of mobile robots equipped with wireless communication device and odor sensors is employed. The experiment is conduct in area of 488cm × 488cm with dynamic odor source in one end. The experiment also used a set of camera to track robots position. The experiement result verifies that PSO is technically sound for real-world odor source localization. In this experime...

PSO Algorithm for Single and Multiple Odor Sources Localization Problems: Progress and Challenge

International Journal on Smart Sensing and Intelligent Systems, 2016

Odor sensing technology in robotic research introduce two research field namely odor recognition ... more Odor sensing technology in robotic research introduce two research field namely odor recognition and odor source localization. Odor source localization research also includes the odor recognition ability with localization method. This paper shows some experiment had been done to localize odor source using single agent and multiple agents. Experiment shows that single agent can’t be used in dynamic environment, hence also can’t be used in real life application. This paper promotes an algorithm known as Particle Swarm Optimization (PSO) to solve these problems. The experiment conducted using PSO shows that PSO able to localize the odor source in the same condition where single agent failed. However, PSO still need to be modified before it can be use widely. This paper shows modification that has been proposed by the authors to enhance it’s ability. The research also has been push to solve multiple odor sources using parallel localization. To verify proposed method, software simulator ...

The STRANDS Project: Long-Term Autonomy in Everyday Environments

IEEE Robotics & Automation Magazine, 2017

hanks to the efforts of the robotics and autonomous systems community, the myriad applications an... more hanks to the efforts of the robotics and autonomous systems community, the myriad applications and capacities of robots are ever increasing. There is increasing demand from end users for autonomous service robots that can operate in real environments for extended periods. In the Spatiotemporal Representations and Activities for Cognitive Control in Long-Term Scenarios (STRANDS) project (http://strandsproject.eu), we are tackling this demand head-on by integrating state-of-the-art artificial intelligence and robotics research into mobile service robots and deploying these systems for long-term in stallations in security and care environments. Our robots have been operational for a combined duration of 104 days over four deployments, autonomously performing end-user-defined tasks and traversing 116 km in the process. In this article, we describe the approach we used to enable long-term autonomous operation in everyday environments and how our robots are able to use their long run times to improve their own performance. Long-Term Autonomy in STRANDS Autonomous robots come in myriad forms and can be used in a range of applications. With these differences, long-term autonomy (LTA) has a variety of meanings. For example, NASA's Opportunity rover has been autonomous for more than ten years on the surface of Mars; wave gliders can automatically monitor stretches of ocean for months at a time; and autonomous cars have completed journeys of thousands of kilometers. In this article, we restrict our contributions to mobile robots operating in everyday, indoor environments, such as offices and hospitals, and capable of performing a variety of The STRANDS Project The STRANDS Project by nick hawes, chris burbridge, ferdian Jovan, lars Kunze, bruno lacerda, lenka mudrová, Jay young, Jeremy wyatt, denise hebesberger, Tobias Körtner, rares ambrus, nils bore, John folkesson, patric Jensfelt, lucas beyer, alexander hermans, bastian leibe, aitor aldoma, Thomas fäulhammer, michael Zillich, markus Vincze, Eris chinellato, muhannad al-omari, paul duckworth, yiannis gatsoulis, david c. hogg, anthony g. cohn, christian dondrup, Jaime pulido fentanes, Tomáš Krajník, João m. santos, Tom duckett, and marc hanheide Long-Term Autonomy in Everyday Environments

Tracking efficiency measurement of dynamic models on geometric particle filter using KLD-resampling

2014 International Conference on Advanced Computer Science and Information System, 2014

Efficiently exploring for human robot interaction: partially observable Poisson processes

Autonomous Robots, Oct 28, 2022

Consider a mobile robot exploring an office building with the aim of observing as much human acti... more Consider a mobile robot exploring an office building with the aim of observing as much human activity as possible over several days. It must learn where and when people are to be found, count the observed activities, and revisit popular places at the right time. In this paper we present a series of Bayesian estimators for the levels of human activity that improve on simple counting. We then show how these estimators can be used to drive efficient exploration for human activities. The estimators arise from modelling the human activity counts as a partially observable Poisson process (POPP). This paper presents novel extensions to POPP for the following cases: (i) the robot's sensors are correlated, (ii) the robot's sensor model, itself built from data, is also unreliable, (iii) both are combined. It also combines the resulting Bayesian estimators with a simple, but effective solution to the exploration-exploitation trade-off faced by the robot in a real deployment. A series of 15 day robot deployments show how our approach boosts the number of human activities observed by 70% relative to a baseline and produces more accurate estimates of the level of human activity in each place and time.

Planning problems in Petri Nets and Fluent Calculus

Fluent Calculus ’ and the work presented in it are my own. I confirm that: This work was done who... more Fluent Calculus ’ and the work presented in it are my own. I confirm that: This work was done wholly or mainly while in candidature for a research degree at Technische Universität Dresden. Where any part of this thesis has previously been submitted for a degree or any other qualification at Technische Universität Dresden or any other institution, this has been clearly stated. Where I have consulted the published work of others, this is always clearly attributed. Where I have quoted from the work of others, the source is always given. With the exception of such quotations, this thesis is entirely my own work. I have acknowledged all main sources of help. Where the thesis is based on work done by myself jointly with others, I have made clear exactly what was done by others and what I have contributed myself.

Efficient Bayesian Methods for Counting Processes in Partially Observable Environments

When sensors that count events are unreliable, the data sets that result cannot be trusted. We ad... more When sensors that count events are unreliable, the data sets that result cannot be trusted. We address this common problem by developing practical Bayesian estimators for a partially observable Poisson process (POPP). Unlike Bayesian estimation for a fully observable Poisson process (FOPP) this is non-trivial, since there is no conjugate density for a POPP and the posterior has a number of elements that grow exponentially in the number of observed intervals. We present two tractable approximations, which we combine in a switching filter. This switching filter enables efficient and accurate estimation of the posterior. We perform a detailed empirical analysis, using both simulated and realworld data.

A Poisson-Spectral Model for Modelling the Spatio-Temporal Patterns in Human Data Observed by a Robot

The efficiency of autonomous robots depends on how well they understand their operational environ... more The efficiency of autonomous robots depends on how well they understand their operational environment. While most of the traditional environment models focus on the spatial representation, long-term mobile robot operation in human populated environments requires that the robots have a basic understanding of human behaviour. We present a framework that allows us to retrieve and represent aggregate human behaviour in large populated environments on extended temporal scales. Our approach, based on combination of time-varying Poisson process models and spectral analysis, efficiently retrieves long-term re-occurring patterns of human activity from robot-gathered observations and uses these patterns to predict human activity and classify locations. The application of our framework on real-world data gathered by a mobile robot operating in an indoor environment for one month indicates that its predictive capabilities outperform traditional temporal modelling methods while being computation...

SOMA : A Framework for Understanding Change in Everyday Environments Using Semantic Object Maps

Understanding change related to the dynamics of people and objects in everyday environments is a ... more Understanding change related to the dynamics of people and objects in everyday environments is a challenging problem. At the same time, it is a key requirement in many applications of autonomous mobile service robots. In this paper we present a novel semantic mapping framework which maps locations of objects, regions of interest, and movements of people over time. Our aim with this framework is twofold: (1) we want to allow robots to reason semantically, spatially, and temporally about their environment, and (2) we want to enable researchers to investigate research questions in the context of long-term scenarios in dynamic environments. Experimental results demonstrate the effectiveness of the framework which was deployed on mobile robot systems in real-world environments over several months.

STRANDS people trajectories - G4S deployment

A collection of detected people trajectories captured using a mobile robot during a six week depl... more

Real-time multisensor people tracking for human-robot spatial interaction

All currently used mobile robot platforms are able to navigate safely through their environment, ... more All currently used mobile robot platforms are able to navigate safely through their environment, avoiding static and dynamic obstacles. However, in human populated environments mere obstacle avoidance is not sufficient to make humans feel comfortable and safe around robots. To this end, a large community is currently producing human-aware navigation approaches to create a more socially acceptable robot behaviour. Amajorbuilding block for all Human-Robot Spatial Interaction is the ability of detecting and tracking humans in the vicinity of the robot. We present a fully integrated people perception framework, designed to run in real-time on a mobile robot. This framework employs detectors based on laser and RGB-D data and a tracking approach able to fuse multiple detectors using different versions of data association and Kalman filtering. The resulting trajectories are transformed into Qualitative Spatial Relations based on a Qualitative Trajectory Calculus, to learn and classify diff...

Advanced Petri Nets and the Fluent Calculus

In this paper we discuss conjunctive planning problems in the context of the fluent calculus and ... more In this paper we discuss conjunctive planning problems in the context of the fluent calculus and Petri nets. We show that both formalisms are equivalent in solving these problems. Thereafter, we extend actions to contain preconditions as well as obstacles. This requires to extend the fluent calculus as well as Petri nets. Again, we show that both extended formalisms are equivalent.

Unsupervised Learning of Qualitative Motion Behaviours by a Mobile Robot

The success of mobile robots, in daily living environments, depends on their capabilities to unde... more The success of mobile robots, in daily living environments, depends on their capabilities to understand human movements and interact in a safe manner. This paper presents a novel unsupervised qualitative-relational framework for learning human motion patterns using a single mobile robot platform. It is capable of learning human motion patterns in real-world environments, in order to predict future behaviours. This previously untackled task is challenging because of the limited field of view provided by a single mobile robot. It is only able to observe one location at any time, resulting in incomplete and partial human detections and trajectories. Central to the success of the presented framework is mapping the detections into an abstract qualitative space, and then characterising motion invariant to exact metric position. This framework was used by a physical robot autonomously patrolling a company's office during a six week deployment. Experimental results from this deployment ...