The difficult task of creating reliable mobility for humanoid robots has been studied for decades. Even though several different walking strategies have been put forth and walking performance has substantially increased, stability still needs to catch up to expectations. Applications for Reinforcement Learning (RL) techniques are constrained by low convergence and ineffective training. This paper develops a new robust and efficient framework based on the Robotis-OP2 humanoid robot combined with a typical trajectory-generating controller and Deep Reinforcement Learning (DRL) to overcome these limitations. This framework consists of optimizing the walking trajectory parameters and posture balancing system. Multi-sensors of the robot are used for parameter optimization. Walking parameters are optimized using the Dueling Double Deep Q Network (D3QN), one of the DRL algorithms, in the Webots simulator. The hip strategy is adopted for the posture balancing system. Experimental studies are carried out in both simulation and real environments with the proposed framework and Robotis-OP2’s walking algorithm. Experimental results show that the robot performs more stable walking with the proposed framework than Robotis-OP2’s walking algorithm. It is thought that the proposed framework will be beneficial for researchers studying in the field of humanoid robot locomotion.

Food drying is a widely used preservation technique; however, achieving high energy efficiency while maintaining product quality remains a significant challenge. This study aims to analyze comprehensive experimental data obtained during the hot-air drying process of the Paşa pear (regional pear) and the system’s autonomous control structure using an explainable artificial intelligence (XAI)-based method. The intelligent drying system, operating for approximately 17.5 h under two temperatures (50 °C and 65 °C) and two air speeds (0.63 m/s and 1.03 m/s), continuously adjusted the temperature and air speed using a PLC-based control mechanism; it ensured stable control throughout the process by monitoring parameters such as product weight, moisture, inlet–outlet temperatures, and air speed in real time. Experimental results showed that drying performance varied significantly with operating conditions, with product mass decreasing from 450 g to 103 g. The innovative aspect of the study is that it obtained quantitative, interpretable rules without discretization by applying the oscillatory chaotic sunflower optimization algorithm (OCSFO) to multidimensional control and process data for the first time. Thanks to its chaotic search mechanism, OCSFO accurately analyzed complex drying dynamics and created rules that achieved over 90% success for high, medium, and low performance classes. The obtained explainable rules clearly demonstrate that drying temperature and air velocity are the dominant determining parameters for drying efficiency, while energy consumption and cabin temperature distribution play a supporting role in distinguishing between efficiency classes. These rules clearly demonstrate how changes in controlled temperature and air velocity, combined with product weight and heat transfer, affect drying performance. Thus, the study offers a robust framework that identifies critical factors affecting drying performance through a transparent artificial intelligence approach that leverages both the autonomous control system and XAI-based rule mining.

In hyperspectral image classification, the most suitable classifier depends on the magnitude and the type of degradation present in the scene; this information, however, cannot be obtained without labels at the stage where the choice has to be made. A two-layer diagnosis computed before classification and without labels is proposed in this study: a severity index reports the magnitude of the degradation, while three scene-derived indicators separate independent, band-dependent, spectrally correlated and striped structures from one another, while classification itself remains supervised. The approach is evaluated on ten benchmark scenes under a leakage-aware spatial protocol. Classifier fragility is found to follow two opposing regimes: tree ensembles are robust to spatially local degradation but fragile to degradation spread across the spectrum, whereas the opposite is observed for the convolutional network. Regime-based classifier selection produces the better classification map in every combination examined. Replacing the spatial protocol with a random pixel split is further shown to inflate accuracy by between 0.162 and 0.249 and to change the best classifier in six of the ten scenes.