School of Aerospace, Transport and Manufacturing (SATM)Selected Masters Theseshttps://dspace.lib.cranfield.ac.uk/handle/1826/87362024-03-28T16:32:26Z2024-03-28T16:32:26ZPassenger spoofing attack for artificial Intelligence-based Mobility-as-a-ServiceChu, Kai-FungGuo, Weisihttps://dspace.lib.cranfield.ac.uk/handle/1826/211082024-03-28T14:29:41Z2024-02-13T00:00:00ZPassenger spoofing attack for artificial Intelligence-based Mobility-as-a-Service
Chu, Kai-Fung; Guo, Weisi
Mobility-as-a-Service (MaaS), a new mobility service model that integrates multiple mobility providers, relies on many data processing technologies to manage multi-modal transport. Artificial Intelligence (AI) is one of the technologies to improve the services matching to passengers based on their implicit experience and preference. However, incorporating AI into MaaS may also introduce loopholes to the system. One may use the loophole in the heterogeneity of passenger experience and preference by falsifying data to prioritize their journey, which jeopardizes the trustworthiness of MaaS. In this paper, we investigate the cyber security risks in MaaS, focusing on the spoofing attack in which malicious passengers are prioritized by falsifying data to gain an advantage in journey planning. The spoofing attack is based on reinforcement learning that learns to reduce passenger satisfaction about the MaaS and its profit by requesting travel with falsifying passenger states. We conduct experiments based on New York City dataset to evaluate the spoofing attack. The experiment results indicate that the attack can reduce about 70% of the profit. By investigating the cyber security risks in MaaS, we could enhance the knowledge and understanding of the risks for building a secure and trustworthy MaaS.
2024-02-13T00:00:00ZFederated reinforcement learning for consumers privacy protection in Mobility-as-a-ServiceChu, Kai-FungGuo, Weisihttps://dspace.lib.cranfield.ac.uk/handle/1826/211072024-03-28T14:13:43Z2024-02-13T00:00:00ZFederated reinforcement learning for consumers privacy protection in Mobility-as-a-Service
Chu, Kai-Fung; Guo, Weisi
Mobility-as-a-Service (MaaS) offers multi-modal transport modes in a single service platform, which requires tremendous data and software support. Among various types of data, consumers' data is vulnerable to the communication channel as it must be transmitted from the consumer end to the MaaS. Consumers put a high priority on the privacy of their data in selecting a service. This motivates the need for a secure information management system for MaaS to protect consumers' information from leakage. In this paper, we propose a federated reinforcement learning (FRL) approach for the information exchange intensive multi-modal journey planning process. The FRL approach protects the information from malicious information thieves by federating the global model training to a local one without sensitive information exchange while maintaining the same solution quality of enhancing MaaS profit and consumer satisfaction. We perform experiments on a test case based on New York City data. The results demonstrate that the FRL approach is effective in the MaaS multi-modal journey planning process. Compared to the baseline approaches, consumer satisfaction and MaaS profit increase by about 12% and 74%, respectively. This pilot study not only provides privacy protection insight into the MaaS multi-modal journey planning but also other privacy-concern applications.
2024-02-13T00:00:00ZL1 adaptive fault‑tolerant control of stratospheric airshipsSouanef, ToufikWhidborne, James F.Liu, Shi Qianhttps://dspace.lib.cranfield.ac.uk/handle/1826/211002024-03-27T13:49:58Z2024-03-26T00:00:00ZL1 adaptive fault‑tolerant control of stratospheric airships
Souanef, Toufik; Whidborne, James F.; Liu, Shi Qian
As the utilization of stratospheric airships becomes more prevalent, ensuring their safe operation becomes crucial. This paper explores the ability of an L1 adaptive controller to maintain fault tolerance in the actuators of a stratospheric airship. L1 adaptive control offers fast adaptation while separating adaptation and robustness. This makes the approach a suitable candidate for fault-tolerant control. The performance of the proposed design is compared to the Linear Quadratic Integral and Adaptive Sliding Mode Backstepping controllers. Simulation results show that the robustness of the airship model against faults is improved with the use of the L1 adaptive controller.
2024-03-26T00:00:00ZAn evaluation of large diameter through-thickness metallic pins in compositesNeale, GeoffreySaaran, VinodhenDahale, MonaliSkordos, Alexandros A.https://dspace.lib.cranfield.ac.uk/handle/1826/210932024-03-27T10:45:20Z2024-03-24T00:00:00ZAn evaluation of large diameter through-thickness metallic pins in composites
Neale, Geoffrey; Saaran, Vinodhen; Dahale, Monali; Skordos, Alexandros A.
There is increasing demand for functional through-thickness reinforcement (TTR) in composites using elements whose geometry exceeds limitations of existing TTR methods like tufting, stitching, and z-pinning. Recently, static insertion of large diameter TTR pins into heated prepreg stacks has proven a feasible and robust reinforcement process capable of providing accurate TTR element placement with low insertion forces and lower tow damage compared with existing methods for similar element sizes (>1mm diameter) like post-cure drilling. Local mechanical performance and failure mechanics of these pinned laminates are reported here. Laminates with a single statically inserted pins (1.2, 1.5, and 2.0 mm) can mostly retain their in-plane integrity alongside a local improvement in mode I delamination toughness in carbon fibre-benzoxazine laminates. Tensile strength is mostly unaffected by the pins resulting from delamination suppression, whereas there is up to a doubling of Young’s modulus. Compressive strength is significantly diminished (up to 42 %) in pinned laminates. Interlaminar toughness is improved, and peak toughness is pushed ahead of the crack as pin diameter increases. The lack of significant deterioration in in-plane tensile properties in pinned laminates produced using static insertion can expand the range and forms of materials that can be inserted compared to existing TTR.
2024-03-24T00:00:00Z