Advanced Double Layered Multi-Agent Systems Based on A3C in Real-Time Path Planning

In this paper, we propose an advanced double layered multi-agent system to reduce learning time, expressing a state space using a 2D grid. This system is based on asynchronous advantage actor-critic systems (A3C) and reduces the state space that agents need to consider by hierarchically expressing a 2D grid space and determining actions. Specifically, the state space is expressed in the upper and lower layers. Based on the learning results using A3C in the lower layer, the upper layer makes decisions without additional learning, and accordingly, the total learning time can be reduced. Our method was verified experimentally using a virtual autonomous surface vehicle simulator. It reduced the learning time required to reach a 90% goal achievement rate by 7.1% compared to the conventional double layered A3C. In addition, the goal achievement by the proposed method was 18.86% higher than that of the traditional double layered A3C over 20,000 learning episodes.

A comparative study of simulation results between PHERB, ADVISOR and AUTONOMIE models

In this research, a new model development and innovation of conventional boat named plug-in hybrid electric recreational boat (PHERB) was introduced to overcome the fuel economy nowadays. This paper focus on the comparison simulation results of PHERB with advanced vehicle simulator (ADVISOR) and an automotive simulation and analysis tool (AUTONOMIE) models in terms of electric machine, energy system storage, propeller and boat model. The model of the PHERB is consequent and applied mathematically in the MATLAB/SIMULINK environment to study its functioning performance. Besides that, fuel economy and emissions of PHERB, ADVISOR and AUTONOMIE model are equated and the pros and cons were discussed in this paper based on simulation result.

myBas driving cycle for Kuala Terengganu city

Driving cycles are series of data points that represent vehicle speed versus time sequenced profile developed for specific road, route, city or certain location. It is widely utilized in the application of vehicle manufacturers, environmentalists and traffic engineers. Since the vehicles are one of the higher air pollution sources, driving cycle is needed to evaluate the fuel consumption and exhaust emissions. The main objectives in this study are to develop and characterize the driving cycle for myBAS in Kuala Terengganu city using established k-means clustering method and to analyse the fuel consumption and emissions using advanced vehicle simulator (ADVISOR). Operation of myBAS offers 7 trunk routes and one feeder route. The research covered on two operation routes of myBAS which is Kuala Terengganu city-feeder and from Kuala Terengganu to Jeti Merang where the speed-time data is collected using on-board measurement method. In general, driving cycle is made up of a few micro-trips, defined as the trip made between two idling periods. These micro-trips cluster by using the k-means clustering method and matrix laboratory software (MATLAB) is used in developing myBAS driving cycle. Typically, developing the driving cycle based on the real-world in resulting improved the fuel economy and emissions of myBAS.

Accelerated testing of super lightweight UHPC waffle deck under heavy vehicle simulator

A super lightweight deck can enhance load rating and functionality of a bridge, especially those identified as structurally deficient. This study was aimed to develop and experimentally validate a novel bridge deck as an ultra-lightweight low-profile waffle slab of ultra-high-performance concrete (UHPC) with either carbon fiber reinforced polymer (CFRP) or high strength steel (HSS) reinforcement. The proposed system lends itself to accelerated bridge construction, rapid deck replacement in bridges with load restrictions, and bridge widening applications without the need to replace girders. Performance and failure modes of the proposed deck were initially assessed through extensive lab experiments and finite element analysis, which together confirmed that the proposed deck panel meets the AASHTO LRFD requirements. The proposed deck system is not susceptible to punching shear of its thin slab and fails in a rather ductile manner. To evaluate its long-term performance, the system was further tested under the dynamic impact of wheel load at the Accelerated Pavement Testing (APT) facility of the Florida Department of Transportation using a Heavy Vehicle Simulator (HVS).