This paper proposes a novel data-driven technique, namely Online Learning for sensor Fault diagnosis Analysis (OLFA), to perform real-time fault analysis for autonomous cars. Considering the non-stationary properties of real-time sensor faults and the mapping relationship between sensors and feature variables, the proposed method decomposes the sensor fault diagnosis analysis problem into an online data stream classification and feature ranking problems. To detect and identify faults, a clustering-based data stream classification approach is developed to continuously capture and classify non-stationary sensor faults for autonomous cars with little intervention from human experts. An effective active learning method is extended and embedded into the proposed framework to minimize the need for prior knowledge about faults and enable the continual learning capability to adapt to and handle the non-stationary properties of sensor faults. Moreover, the proposed framework addresses the parameter optimization issue of existing machine learning based fault analysis techniques and employs feature ranking analysis to systematically analyze the possible source(s) of sensor faults. CAR Learning to Act (CARLA), a well-known realistic autonomous driving simulator, is used as the benchmark to perform the sensor fault injection and online data stream collection to evaluate the efficacy of OLFA. Analysis of the collected faulty datasets and experimental results, and comparison between OLFA and several state-of-the-art clustering-based approaches for fault classification, demonstrated the efficacy of the proposed framework in the domain of autonomous cars.

Streaming feature selection (SFS) is emerging as a key research direction which addresses the non-stationary property of feature streams when the sample size is fixed. Most existing SFS techniques are supervised methods, and ignore the label scarcity. Real-world datasets are typically unlabeled and the labeling costs are expensive. Although some unsupervised SFS approaches are proposed, these approaches are either limited to the homogeneous feature types or require substantial computational complexity. To address these problems, we propose an online unsupervised feature selection framework using dynamic feature clustering in this paper. We derived a recursive density lower bound to estimate the density distribution of feature streams and developed a density-based dynamic clustering method to perform the online feature stream clustering for exploring feature redundancy. An unsupervised online feature relevance maximization and redundancy minimization strategy is introduced to extract a subset of important features with low redundancy from the feature stream. Experimental results on thirteen well-known benchmark datasets and comparison studies with seven state-of-the-art supervised SFS methods demonstrate that the proposed unsupervised method provides statistically comparable performance with the supervised SFS techniques while the label information is unknown.

We present the design and implementation of a prototype of Bangla voice controlled robotic motion based on an efficient and unique algorithm for accurately detecting Bangla voice commands. For the detection of Bangla voice commands to control the movement, our developed speech detection system incorporates Mel-Frequency Cepstrum Coefficients based speech feature extraction mechanism and a feature matching technique using Vector Quantization. Our proposed Bangla voice controlled robotic motion can be very effective in rescue operation and disaster management system where a rescuer can dynamically give the rescue robot voiced instructions instead of being stagnant in a place for its controlling. Additional features including dynamic path generating algorithm for the traversed path, capability to accurately follow a specific map to reach a certain destination, detection of different gases and dynamic obstacle avoidance as well as precise position and movement control system add to the flexibility and dimension of the robotic motion. Apart from its possible application in industrial automation, Bangla voice controlled robotic motion can also encourage the view of a robotic machine as a partner rather than a tool by providing a smooth, easy and flexible mean of self-reliant movement for the physically challenged people speaking Bangla language throughout the world.

In this paper, we develop a data-driven performance-based evaluation framework for an electric vertical takeoff and landing aircraft (eVTOL) in the context of Urban Air Mobility (UAM) applications. First, a two-stage comprehensive simulation framework is developed to generate a benchmark database for the performance evaluation of both UAS Traffic Management (UTM) algorithms and high-fidelity eVTOL dynamical models. In the developed simulation framework, we implement UTM algorithms and incorporate real-world constraints, e.g., vertiport infrastructures and different wind conditions. From the developed simulation framework, we generate 1,213,010 flight profiles. These flight profiles are used in a model based eVTOL performance evaluation tool as inputs to compute the physical performance of 3 types of eVTOLs. Due to the high computational cost of model-based eVTOL performance evaluation approaches, a clustering-based sampling procedure is employed to reduce the redundancy in the generated flight profiles and utilize the re-sampled flight profiles to form an eVTOL performance analysis dataset. We then train and compare several machine learning models on the eVTOL performance analysis dataset to predict: performance variables-flight conditions, aerodynamic coefficients, aircraft electronics, electric motor and propeller efficiencies. Finally, we deploy the proposed data-driven models in the framework and reduce the eVTOL performance inference time to real-time. The implementation of the proposed framework can be found on GitHub: https://github.com/mrinmoysarkar/eVTOL_performance_evaluation.git