Events
19 October 2023
Doctoral Researcher Seminar
Presented By Speaker: Adam McKinley (CE), Supervisors: Professor Wen-Feng, Lin and Dr Simon Kondrat Title: Development of Ni based catalysts for hydrogen evolution reaction (HER) Tim Glover (AAE), Supervisors: Professor Wen-Hua Chen and Dr Cunjia Liu Title: Dual Control Inspired Active Sensing for Bearing-Only Target Tracking
- 12:00 - 13:00
- DAV1100 (Sir David Davies Building) - 探花精选 Campus
About this event
Speaker: Adam McKinley (CE)
Supervisors: Professor Wen-Feng Lin and Dr Simon Kondrat
Title: Development of Ni based catalysts for hydrogen evolution reaction (HER)
Abstract
Over the past decade there has been a significant escalation in the interest for the utilisation of an alternative fuel source that is sustainable, storable and economically viable. Hydrogen has the highest gravimetric energy density when compared to other fuels available. Currently, expensive platinum group metals (PGMs) are the “gold standard” for electrocatalysts involved in hydrogen production.
This presentation will analyse and compare the catalytic activity of multiple Ni based catalysts, including a Fe coated Ni pellet as well as Ni foam. With the main aim being to improve the economic viability of water electrolysis.
Speaker: Tim Glover (AAE)
Supervisors: Professor Wen-Hua Chen and Dr Cunjia Liu
Title: Dual Control Inspired Active Sensing for Bearing-Only Target Tracking
Abstract
This presentation outlines an active sensing algorithm for target tracking that combines task based and information-based approaches through the dual control for exploitation and exploration (DCEE) concept. A mobile sensor platform with a limited field-of-view, performing bearing-only target tracking, is controlled via the DCEE based cost function using a Monte Carlo tree search (MCTS) framework for nonmyopic, online, decision making. The DCEE observer control method is benchmarked against the information theoretic Rényi divergence. Rényi divergence performs marginally better when considering target existence estimation, but spatial results clearly demonstrate that our formulation outperforms the benchmark algorithm with improved target localisation.