Abstract: The remaining useful life (RUL) prediction is crucial for maintaining aircraft engines' safety and operational efficiency, requiring precise forecasting in their complex and dynamic environments. This paper presents a novel approach for predicting RUL of aircraft engines using an advanced long short-term memory (LSTM) network. The novelty of this approach lies in the integration of a specially designed loss function and the introduction of a new evaluation metric called the Average Safe Underestimation Error (ASUE). The evaluation scheme utilizes the comprehensive C-MAPSS dataset from NASA, which includes detailed run-to-failure operational data of aircraft engines. The LSTM network, specifically adapted for time-series analysis in RUL prediction, underwent a meticulous preprocessing phase to tailor the dataset for effective LSTM analysis. The network was then trained and optimized to predict RUL with high accuracy. A key challenge in RUL prediction is balancing the risk of overestimating RUL, which can lead to unexpected failures, and under predicting it, which can result in unnecessary maintenance costs. To address this, a modified loss function is incorporated in the LSTM model that specifically penalizes overestimation, thus reducing the likelihood of unexpected engine failures. Furthermore, the ASUE metric is introduced to quantitatively evaluate the model's performance in maintaining a delicate balance between ensuring operational safety and minimizing maintenance costs. The proposed approach, combining the modified loss function and the ASUE metric, steers the LSTM model towards more accurate, reliable, and cost-effective RUL predictions. This enhances safety and operational efficiency in aviation maintenance, contributing significantly to the field of predictive maintenance in the aerospace industry. The study demonstrates that our methodology not only improves the accuracy of RUL predictions but also plays a vital role in aligning them with economic considerations and safety requirements.

Abstract: Grasp coordination is crucial for performing cooperative object manipulation tasks. Planning cooperative grasps among a group of decentralized robots is a new line of coordination problem that requires robots to simultaneously handle complex and large shaped and sized objects, a new level of difficulty that is rarely addressed in the literature. In this paper, we propose grasp coordination approaches for a decentralized group of robots facing explicit communication and sensing limitations. In particular, a scenario where robots with incomplete knowledge about each other's embodiments further lose the ability to 1. observe others' grasps occluded by the object's shape 2. exchange direct messages due to potential communication degradation resulting from the real-time planning and execution constraints. To tackle such a scenario, we introduce two baseline and two probabilistic approaches that are specifically designed for strict decentralization. The approaches analyze cooperative grasps using traditional grasp quality metrics and estimate cooperative grasps based on the assigned robot's priority. Simulation experiments demonstrate that the probabilistic approaches exhibit superior performance over the baseline approaches, reaching performance close to optimal for both homogeneous and heterogeneous groups. These approaches provide solutions to simulated multi-robot grasp coordination scenarios that have the potential to translate to real-world environments such as logistics, manufacturing, and services.

Abstract: Cooperative manipulation has become an important task in both industrial and domestic settings. Control of dexterous robotic systems to assist such task is challenging. In this paper, we first investigate three impedance-based control approaches for robotic assistance in cooperative manipulation. In particular, we study the level of assistance and cooperation from each approach by performing physical human-robot interaction (pHRI) experiments. Modern robotic manipulators are equipped with sophisticated robot hands to perform dexterous manipulation. The multi-fingered hand-arm cooperation with a human partner can be improved by adding admittance or impedance control to the finger part sensing interaction forces. The conventional way of using force/torque sensors at the wrist has limitations in distinguishing internal and external forces as well as relating the point of interaction. Tactile sensing can be useful in such a system to estimate the forces exerted by the human user during the interaction as well as to distinguish the applied contact force. Second, detection of human intended forces is demonstrated with a tactile based multi-fingered robot hand. Finally, a novel design of an assistance system that incorporates tactile sensing and chosen assistance approach modeled as a virtual tool in admittance control is presented for dexterous robot-human cooperative object manipulation.