Human Activity Recognition (HAR) is vital for understanding human behavior and improving people'slives. The overall goal of a reliable HAR system is to automatically analyze and comprehend humanactions based on the data collected from various sensors and sources, and different environments. In thisresearch, we utilize a Frequency Modulated Continuous Wave radar sensor. Radar-based HAR has severaladvantages over camera-based HAR. It is less affected by external environmental factors, such as lightingconditions, smoke, or dust, making it easily deployable and cost-effective. Furthermore, compared to thecamera-based HAR which may raise privacy concerns, radar-HAR is less invasive as it does not capturevisual details of the participants.In recent years, machine learning algorithms have made significant progress in analyzing big data, andHAR is one of the key areas where machine learning algorithms have been applied successfully. Much ofthis success is due to supervised learning which relies heavily on the availability of large labeled datasets.However, creating labeled datasets is a time-consuming and expensive process that requires significanthuman effort and domain expertise. As a result, this makes supervised learning infeasible for uniqueor niche domains with limited data availability, such as radar-HAR. Additionally, supervised learningalgorithms may not perform well on unseen data if there is a significant difference between the trainingand testing datasets, known as the “domain-shift” or “dataset bias” problem.In this thesis, our primary goal is to focus on the generalizability aspect of the deep learning-basedmodels in the presence of domain-shift for an indoor radar-HAR application. In this context, radar targettracking-based auxiliary features and the preprocessing steps in the radar data cube are proposed. Thetracking-based features provide the dynamic context of the participants in an indoor environment. Atthe same time, the proposed preprocessing steps in the radar data cube are driven by the Doppler &range energy dispersion-based profiles of the participant's micro-motion. A novel Multi-View CNN-LSTM-based multi-model approach is proposed, which efficiently combines the complementary holisticview of the participants, given by the dynamic auxiliary contextual features, with the energy dispersion-based spatiotemporal features from the preprocessed radar data cubes. Moreover, to facilitate robustand class-agnostic feature extraction an unsupervised Convolutional Auto-encoder-based training andmodel initialization step is proposed, followed by the supervised training and fine-tuning steps. Lastly, toaddress the domain-shift problem, the proposed robust model training methodology is extended with theCORrelation ALignment (CORAL)-based Multi-View Unsupervised Domain Adaptation-based modeladaptation step.