Page Not Found
Page not found. Your pixels are in another canvas.
Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
My academic webpage
About me
Page not in menu
This is a page not in th emain menu
Posts
portfolio
publications
Topological Robot Localization in a Pipe Network
Published in UKRAS20 Conference Robots into the real world Proceedings, 2020
To do robot localization, the robot state to be estimate has to be defined. This paper proposes the use of a discrete (topological) robot state definition for robots in pipe networks, as an alternative to the traditional continuous (metric) approach which is not well suited to this constrained environment.
Acoustic echo-localization for pipe inspection robots
Published in 2020 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI), 2020
This paper proposes the use of acoustic echo sensing for robots in pipe networks, and develops an algorithm for using this sensing in localization. This sensing approach increases the robot's perspective compared to the use of typical sensors such as vision, which are limited in range in this environment.
Topological robot localization in a large-scale water pipe network
Published in Towards Autonomous Robotic Systems: 21st Annual Conference, TAROS 2020, Nottingham, UK, September 16, 2020, Proceedings 21, 2020
This paper evaluates the use of a topological robot state definition for robots in pipe networks proposed in my previous work. The sources of uncertainty which cause an increase in estimate error are identified. Alternative state definitions are propsed which result in different values of efficiency, in terms of estimate accuracy relative to the required computational time. On reflection, the attempt to improve efficiency by using the proposed state definition went too far away from the typical state definition; this motivated further work to find a compromise between the two approaches.
Robot Localization in Water Pipes Using Acoustic Signals and Pose Graph Optimization
Published in Sensors, 2020
This paper continues work done by a colleague, which uses a hydrophone sensor to measure a physical property of a pipe. This varies when the measurement is made by a robot at different points along the pipe, and therefore can be used to localize a robot. This work develops this localization approach by applying the sensing approach to pose-graph optimization. On reflection, this work shows that this sensing could be useful in the case where sensors like vision are unavailable or unsuitable due to the feature-sparseness of the pipe. However, like the use of vision, this approach suffers from a fundamental limitation on its perspective, which motivates further work to overcome this.
Robust efficient localization of robots in pipe networks using a particle filter for hybrid metric-topological space
Published in 2021 European Conference on Mobile Robots (ECMR), 2021
This paper proposes a hybrid continuous-discrete (metric-topological) robot state definition for robots in pipe networks, overcoming the limitations of the approach proposed in my previous work, and finding a middle ground between that and alternatives presented in existing literature
Robot localization in a pipe network using a particle filter with error detection and recovery in a hybrid metric-topological space
Published in 2021 IEEE International Conference on Multisensor Fusion and Integration (MFI 2021), 2021
This paper shows some improvements to an algorithm based on a hybrid continuous-discrete (metric-topological) robot state definition. Specifically, it presents an approach to the challenging problem of error detection in this environment in the case of limited sensor information, and it presents an approach to error recovery which uses the constraints of the environment to its advantage.
Acoustic Echo Sensing for Robot Localization in Buried Pipe Networks
Posted to TechrXiv, 2023
This paper further develops the use of acoustic echoes for robot localization in a pipe. The localization algorithm is improved to facilitate measurements in more realistic and challenging environments than in previous work. The sensing, perception, and localization approaches are all evaluated in more thorough experiments than in previous work. This is likely as far as localization can be developed for this environment using only acoustic sensing; future work would need further experimental validation in a field robotics context, and need to combine information from multiple sensors.
Acoustic Echo Sensing for Robot Localization in Buried Pipe Networks
Published in IEEE Sensors, 2024
This paper further develops the use of acoustic echoes for robot localization in a pipe. The localization algorithm is improved to facilitate measurements in more realistic and challenging environments than in previous work. The sensing, perception, and localization approaches are all evaluated in more thorough experiments than in previous work. This is likely as far as localization can be developed for this environment using only acoustic sensing; future work would need further experimental validation in a field robotics context, and need to combine information from multiple sensors.
Hybrid Metric‐Topological Localization for Robots in Pipe Networks
Published in Journal of Field Robotics, 2024
This paper further develops localization in a large-scale pipe network. This approach could be applied to any network environment, but is especially suited to the limited sensing available for robots in pipe networks. The approach uses a hybrid metric-topological state space, where the state to be estimated contains both continuous and discrete variables. Compared to our previous results, the estimation approach is more robust to most sources of input uncertainty and requires lower computation.