Photovoltaic (PV) array fault diagnosis is important because it helps reduce energy and revenue losses to PV system operators. It also reduces fire hazards and electric shocks caused by PV array faults. As a result, many machine-learning-based fault diagnosis techniques have been proposed in recent times. Although the fault diagnosis accuracies associated with these techniques have been impressive, most machine learning algorithms rely on manual feature extraction, which is time consuming, expensive, and diagnostic expertise exacting. To address the problem of manual feature extraction, this paper proposes a new PV array fault diagnosis technique capable of automatically extracting features from raw data for PV array fault classification. The proposed technique utilizes long short-term memory (LSTM) networks, which is a deep learning (DL) algorithm, for feature extraction. The extracted features feed into a softmax regression classifier for fault diagnosis. The proposed technique exhibits high fault diagnosis accuracies on both noisy and noiseless data. Additionally, the results of the proposed technique compares favorably with those of other techniques. It can, therefore, be inferred from the results that the proposed fault diagnosis technique offers an effective approach to automatically extract useful features from raw data and thus remove the need for manual feature extraction.

Safe and smooth mobile robot navigation through cluttered environment from the initial position to goal with optimal path is required to achieve intelligent autonomous ground vehicles. There are countless research contributions from researchers aiming at finding solution to autonomous mobile robot path planning problems. This paper presents an overview of nature-inspired, conventional, and hybrid path planning strategies employed by researchers over the years for mobile robot path planning problem. The main strengths and challenges of path planning methods employed by researchers were identified and discussed. Future directions for path planning research is given. The results of this paper can significantly enhance how effective path planning methods could be employed and implemented to achieve real-time intelligent autonomous ground vehicles.

This paper presents a novel path planning method for mobile robots in complex and dynamic environments using voronoi diagram (VD) and computation geometry technique (CGT) termed, VD-CGT. An algorithm to categorize moving obstacles based on their positions, velocities, distances, and directions to ascertain their collision threat level, and possible replanning decision is introduced. The initial path computation is done using morphological dilation, VD, A-star and cubic spline algorithms. Instead of considering the entire map of the environment, CGT is used to compute a small rectangular region estimated to enclose a detected collision-threat obstacle and the current position of the robot. Roadmap is computed in the geometrical shape using VD and nodes are added to the initial roadmap nodes to compute a new path for replanning. To avoid increasing time and space requirements, these nodes are discarded before subsequent replanning is done. Results indicate better path replanning performance in complex and dynamic environments in terms of success path computation rate, path cost, time, and number of replanning computations compared with other five popular related path planning approaches. The proposed method is efficient. It computes safe and shortest replan path to goal with low computation time requirement. Unnecessary replanning computations are avoided which aid in reducing time and distance to get to goal. With the performance results, the proposed method is a promising method to achieving safe, less path cost, and time in path replanning computations in complex and dynamic environments.