Given the similar response scenarios between CBRN and Hazmat in peacetime, this research aims to look at areas of integration and delineation between both response forces during a C,B and R incident. Singapore will be used as the basis of comparison when doing the comparative case study, though the global overall CBRN and HazMat scenes will still be studied closely for lessons learnt. The following questions will guide the research:

• What are the key similarities and differences between CBRN and HazMat?
• With peacetime Singapore as the context, what are some of the capability development and operational issues that should be integrated or delineated in the field of CBRN and HazMat?
• What are some of the lessons learnt from CBRN and HazMat incidents worldwide? Which of these can be adopted by Singapore (and which of these should Singapore avoid)?
• How can the general public, private sector, military and government organisations work together to streamline areas in both CBRN and HazMat responses?

Polymer-Bonded Explosive (PBX) ages and degrades over time when exposed to a wide range of environmental conditions during its service life. These often lead to material property changes and can significantly impact PBX’s safety and performance. The accelerated ageing technique, used to predict the shelf life of the munition, is often carried out at temperatures at or above 60 °C to decrease time frame and cost. It is well known that different materials might behave differently at evaluated temperatures, thus altering the energetics’ properties. This would lead to inaccurate predictions on the shelf-life assessment. Therefore, it is essential to understand the actual natural degradation phenomenon to provide additional assurance on the predicted shelf life.

In every portion of a product’s life cycle, system failures can occur. The DOD and defense contractors use failure analysis methods such as Failure Modes and Effects Analysis (FMEA) and Fault Tree Analysis (FTA), which have successfully reduced maintenance downtime and lowered life cycle costs as a result of reducing failures during system operations. However, challenges can hinder the effectiveness of failure analysis and have not been fully addressed due to limitations of current failure analysis and root cause analysis methods. This thesis addresses two of the challenges: (1) incorrectly classified system failures and (2) lack of failure traceability in the system life cycle. A methodology is proposed that incorporates some elements of both FMEA and FTA into a digital twin (DT) by employing the MBSE tool Magic Systems of Systems Architecture (MSOSA) to aid in failure analysis. An MBSE model simulation using an example of an engine is presented to demonstrate the efficacy of the methodology. The results are shown in terms of system operational availability. This research concludes that the approach has potential in emulating the real system’s behavior and can offer a more accurate estimated result. However, due to semantic errors in modeling in MSOSA, the model was unable to generate accurate results as intended. The model’s semantic errors must be addressed before it can be validated by comparing the results from MBSE model and the corresponding real-world system.

Existing counter-unmanned aerial system (C-UAS) defensive mechanisms rely heavily on radio frequency (RF) jamming techniques that require a large amount of energy to operate. The effects of RF jamming result in undesirable consequences, such as the jamming of other nearby friendly radio devices as well as the increase in RF footprint for local operators. Current cybersecurity analysis of commercial off-the-shelf (COTS) UASs have revealed multiple vulnerabilities that give rise to opportunities to conduct C-UAS operations in the cyber domain. This is achieved by performing cyber-attacks on adversarial UASs through hijacking the device-specific communication’s link on a narrow RF band and without the need for broad-spectrum RF energy bursts during C-UAS operations, which can result in lower energy usage to accomplish the same outcome. This article validates the cyber-attack C-UAS (CyC-UAS) concept through reviewing recent C-UAS operational experimental scenarios and conducting analysis on the collected data. Then, a simulation model of a defense facility is constructed to analyze and validate specific mission scenarios of interest and several proposed concepts of operation. A comparison of the energy requirements between CyC-UAS and existing C-UAS techniques is performed to assess energy efficiency and trade-offs of different C-UAS approaches. In this article, the comparison of energy requirements between the CyC-UAS prototype and existing C-UAS products that utilize RF jamming methods reveals that CyC-UAS achieves significant energy savings while not affecting other telecommunication devices operating at the same frequencies. While both the C-UAS techniques adopt the denial-of-service strategy, the CyC-UAS is able to achieve the same mission by consuming much less energy. Therefore, the CyC-UAS concept shows promise as a new, lower energy, and lower collateral damage approach to defending against UAS.

In recent years, militaries have strengthened efforts to integrate unmanned technologies to improve manned-unmanned teaming (MUM-T) capabilities. As some countries’ fighting-age populations are decreasing, militaries are turning to readily available, cost efficient, and sophisticated unmanned technologies. MUM-T holds great potential not only to alleviate manpower shortages in militaries, but also to improve combat capabilities. This thesis studies the effectiveness of MUM-T at the frontline, down to infantry teams supporting offensive operations in urban terrain. An agent-based simulation is used to model a MUM-T combat operation with and without an unmanned ground vehicle (UGV) to support an infantry company. An analysis was conducted on more than 76,800 simulated battles. It was observed that MUM-T concepts could dramatically increase combat effectiveness, as assessed by increased enemy casualties. The UGV reloading time, weapon accuracy, and own force structure were also observed to significantly impact the infantry’s lethality and survivability. This analysis concludes that implementation of MUM-T at the small-unit tactical level has great potential to enhance overall combat performance. Moving forward, combat models could be integrated into future military exercises such that the findings from simulations can be verified and validated.

Many armed forces are becoming network-centric and highly interconnected. This transformation, along with decentralized decision making, has been enabled by technological advancements in the digital battlefield. As the battlefield evolves and missions require units to be mobile and support numerous tactical capabilities, the current concept of deploying static radio-relay nodes to extend the range of communication may no longer be suitable. Hence, this thesis aims to design an operational concept using unmanned aerial systems such as aerostats and tactical drones to provide beyond line-of-sight communication for tactical forces while overcoming the limitations in a GPS-denied environment. The proposed concept is divided into three different phases to assess operational and communication system needs. Given Federal Communications Commission regulations that set the maximum effective isotropic radiated power in the Industrial, Scientific, and Medical band at 36 dBm. The maximum communication range between two nodes can be studied using the Friis propagation equation. In addition, Simulink software is used to study the effective application throughput with respect to distance. From the analysis, IEEE 802.11ax can provide a higher data throughput and support both 2.4 GHz and 5.0 GHz frequency band. Using a simulated environment and operational scenario. the estimated number of aerial systems required to provide communication coverage for a 50 km by 50 km area is determined.

Accurate tracks and targeting are key to providing decision-makers with the confidence to execute their mission. Increasingly, multiple Intelligence, Surveillance, Reconnaissance (ISR) assets across different intelligence sources are being used to increase the accuracy of track location, resulting in the need to develop methods to exploit heterogenous sensor data streams for better target state estimation. One of the algorithms commonly used for target state estimation is the Kalman Filter (KF) algorithm; the algorithm performs well if its covariance matrices are accurate approximation of the uncertainty in sensor measurement. Our research complements the Artificial Intelligence (AI)/Machine Learning (ML) efforts the Navy is taking stride in by quantitatively assessing the potential of using a ML model to predict sensor measurement noise for KF state estimation. We used a computer simulation software to generate sensor tracks of a single target and trained a neural network to predict sensor error. The hybrid model (ML-KF) was able to outperform our baseline KF model that uses normalized sensor errors, by approximately 20% in target position estimation. Further research in enhancing the ML-KF model with external environment variable as inputs could potentially create an adaptative state estimation system that is capable to operate in varied environment settings.

Navigation systems of unmanned aircraft systems (UAS) are heavily dependent on the availability of Global Positioning Systems (GPS) or other Global Navigation Satellite Systems (GNSS). Although inertial navigation systems (INS) can provide position and velocity of an aircraft based on acceleration measurements, the information degrades over time and reduces the capability of the system. In a GPS-denied environment, a UAS must utilize alternative sensor sources for navigating. This thesis presents preliminary evaluation results on the usage of onboard down-looking electro-optical sensors and image matching techniques to assist in GPS-free navigation of aerial platforms. Following the presentation of the fundamental mathematics behind the proposed concept, the thesis analyzes the key results from three flight campaign experiments that use different sets of sensors to collect data. Each of the flight experiments explores different sensor setups, assesses a variety of image processing methods, looks at different terrain environments, and reveals limitations related to the proposed approach. In addition, an attempt to incorporate navigational aid solutions into a navigation system using a Kalman filter is demonstrated. The thesis concludes with recommendations for future research on developing an integrated navigation system that relies on inertial measurement unit data complemented by the positional fixes from the image-matching technique.

The motivation of implementing the single naval fuel was to increase the interoperability between different platforms, simplifying logistics procurement, storage requirement and the smooth transportation of fuel. Streamlining of the JP5 fuel had its limitations due to the reasons covered in the background. Alternative fuel studies have been an ongoing for at least the past two decades, especially for the use of hydrogen fuel technology onboard a variety of land and air platforms. Hydrogen fuel technology has since been operational in ground and air platforms such as the Stalker, Blackjack unmanned aerial vehicles and reckless unmanned terrain vehicles. There are several prototypes for commercial shipping liners and sail boats available currently, but the hydrogen fuel technology is yet to be fully developed for military vessels.

This thesis focuses on using liquid hydrogen (LH2) technologies as a single naval fuel onboard a vessel. The case study in this thesis is a 1000 twenty-foot equivalent unit (TEU) catamaran container vessel (CV) as a proxy to the United States (U.S.) Navy’s Spearhead Class vessel. These two vessels are both built with catamaran hulls and have similar tonnage. The case study was assumed to have travelled a distance of 4838 nm and a travelling time of 194 hrs. Design replacements and modifications were proposed in order to meet the case study energy requirements following which RetScreen and Microsoft Excel were used to evaluate this system based over a 20-year project life. Overall, the proposed hydrogen fuel system as a single naval fuel is financially unfeasible as the economic profits do not look promising when compared to the estimated required investment at the start. Further investigation is recommended to determine whether the proposed hydrogen fuel system mentioned in this thesis can combined with other renewable technologies as they become available. Actual implementation on a military vessel may not be practical now due to the high initial costs.

Satellite imaging performance can degrade due to optical aberrations. To maximize a satellite’s imaging output over its useful lifespan, deep learning presents a cost-effective and efficient alternative to traditional adaptive optics and blind deconvolution for deblurring satellite images. This is because deep learning is essentially a post-processing technique that relies on algorithms and a large dataset. This research focuses on applying deep learning algorithms based on the UNET Convolutional Neural Network, which is widely used in the bio-medical imaging field, to deblur optically aberrated satellite imagery. The XVIEW dataset, which is composed of images taken by the Worldview-3 satellite, is used. The XVIEW images are then simulated with optical aberrations (defocus and spherical) using Zernike polynomials. The blurred images are subsequently deblurred with UNET and UNET variants (UNET++ and UNET3+) before final performance evaluation with various image quality metrics. The results showed that (1) UNET algorithms can effectively deblur optically aberrated satellite images, and (2) UNET3+ modified with additional convolutional layers (deep-UNET3+) provided the best deblurring performance. Based on the positive results, this thesis recommends that the UNET algorithm be applied on actual field cases of optically aberrated satellite imagery and be further developed to perform better even on super-resolution applications.