The localization of coal mine robots (CMRs) serves as the foundation for intelligent mines. Despite various approaches proposed by academia and industry, existing range-based localization methods generally encounter the non-line-ofsight (NLOS) problem, leading to severe accuracy degradation in complicated underground mines. Facing the above challenge, this paper proposes an edge-assisted cross-domain NLOS localization (CrossDNL) framework based on semi-supervised learning. Specifically, we analyze the channel impulse responses (CIRs) of ultra-wideband (UWB) to identify the multi-channel NLOS conditions and further mitigate ranging errors via the deep neural network (DNN). To achieve reliable localization across various scenarios, CrossDNL adopts an edge-assisted semi-supervised learning architecture that enables it to optimize DNN models by self-training on the edge server and achieve the real-time service on the CMR. Besides, we propose a range-based error state Kalman filter (ESKF) scheme to improve the localization performance. We implement CrossDNL on the CMR and the edge server and evaluate it in various scenarios. Results show that CrossDNL can achieve a 14cm localization error in real time, outperforming the latest solutions by more than 20%.

In software reliability assessment, one problem of interest is how to minimize the variance of reliability estimator, which is often considered as an optimization goal. The basic idea is that an estimator with lower variance makes the estimates more predictable and accurate. Adaptive Testing (AT) is an online testing strategy, which can be adopted to minimize the variance of software reliability estimator. In order to reduce the computational overhead of decision-making, the implemented AT strategy in practice deviates from its theoretical design that guarantees AT's local optimality. This work aims to investigate the asymptotic behavior of AT to improve its global performance without losing the local optimality. To this end, a new AT strategy named Adaptive Testing with Gradient Descent method (AT-GD) is proposed. Theoretical analysis indicates that AT-GD, a locally optimal testing strategy, converges to the globally optimal solution as the assessment process proceeds. Simulation and experiments are set up to validate AT-GD's effectiveness and efficiency. Besides, sensitivity analysis of AT-GD is also conducted in this study.

Random testing (RT) and subdomain testing are two major software testing strategies. Their simplicity makes them likely the most efficient testing strategies with respect to the time required for test case selection. However, the disadvantage of RT is its defect detection effectiveness. Adaptive testing (AT) is a feedback-based software testing strategy that has been shown to be more effective than RT and partition testing (PT). However, a major concern in the application of AT is its complexity and computational cost for test case selection. In this paper, we propose a hybrid approach that uses AT and random partition testing (RPT) in an alternating manner. The motivation for this approach is that both strategies are employed such that the underlying computational complexity of AT is reduced by introducing RPT into the testing process without affecting the defect detection effectiveness. A case study with seven real-life subject programs is presented. The experimental results demonstrate that this novel strategy considerably reduces the computational overhead of the original AT strategy but still outperforms the pure RT strategy and PT strategy in terms of the number of test cases used to detect and remove a given number of defects. In addition, a sensitivity analysis is conducted to validate the robustness of our strategy.

Test case prioritization (TCP) techniques aim to schedule the order of regression test suite to maximize some properties, such as early fault detection. In order to measure the abilities of different TCP techniques for early fault detection, a metric named average percentage of faults detected (APFD) is widely adopted. In this paper, we analyze the metric APFD and explore the gain of measuring TCP techniques from a control theory viewpoint. Based on that, we propose a generalized metric for TCP. This new metric focuses on the gain of defining early fault detection and measuring TCP techniques for various needs in different evaluation scenarios. By adopting this new metric, not only flexibility can be guaranteed, but also explicit physical significance for the metric will be provided before evaluation.