Search

Machine learning interatomic potentials have become a powerful tool to achieve molecular dynamics (MD) simulations with the accuracy of ab initio methods while beyond their length and timescale limitations. Here, we develop an efficient moment tensor potential (MTP) for high-entropy diborides (HEBs) based on unary and binary diborides with Ti-V-Cr-Zr-Nb-Mo-Hf-Ta-W principal elements. Notably, the trained MTP exhibits exceptional generalization across both equimolar and non-equimolar HEBs, with testing errors in energy and force of 2.6 meV/atom and 155 meV/{\AA} for equimolar HEBs, and 3.7 meV/atom and 172 meV/{\AA} for non-equimolar HEBs, respectively, indicating its remarkable accuracy and transferability. The reliability of the established MTP is further confirmed by a comparative analysis with first-principles calculations, where our MTP accurately reproduces the structural and mechanical properties of various HEBs. The work presents a significant advancement in the simulation of high-entropy ceramics with enhanced efficiency and accuracy., Comment: 17 pages, 3 figures

Molecular dynamics (MD) simulations are considered an efficient and low-cost means to develop remarkable properties of high-entropy ceramics with vast composition space, yet the lack of general potentials severely limits their applications. Herein, taking high-entropy carbides (HECs) as the model, we propose a strategy to efficiently construct a general neuroevolution potential (NEP) with broad compositional applicability for HECs. Specifically, the small dataset comprising unary and binary carbides with ten transition metal principal elements is first identified as the most efficient choice to train the general NEP for HECs, and then a highly accurate and transferable NEP is constructed. Further MD predictions on structural, mechanical, and thermal properties of HECs using the established NEP show good agreement with the results of first-principles calculations and experimental measurements, validating the accuracy, generalization, and reliability of our developed NEP. Our work provides an efficient solution to accelerating the MD simulations searching for high-entropy ceramics with desirable properties., Comment: 28 pages, 6 figures

Sampling-based motion planning (SBMP) algorithms are renowned for their robust global search capabilities. However, the inherent randomness in their sampling mechanisms often result in inconsistent path quality and limited search efficiency. In response to these challenges, this work proposes a novel deep learning-based motion planning framework, named Transformer-Enhanced Motion Planner (TEMP), which synergizes an Environmental Information Semantic Encoder (EISE) with a Motion Planning Transformer (MPT). EISE converts environmental data into semantic environmental information (SEI), providing MPT with an enriched environmental comprehension. MPT leverages an attention mechanism to dynamically recalibrate its focus on SEI, task objectives, and historical planning data, refining the sampling node generation. To demonstrate the capabilities of TEMP, we train our model using a dataset comprised of planning results produced by the RRT*. EISE and MPT are collaboratively trained, enabling EISE to autonomously learn and extract patterns from environmental data, thereby forming semantic representations that MPT could more effectively interpret and utilize for motion planning. Subsequently, we conducted a systematic evaluation of TEMP's efficacy across diverse task dimensions, which demonstrates that TEMP achieves exceptional performance metrics and a heightened degree of generalizability compared to state-of-the-art SBMPs., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

This paper presents three open-source reinforcement learning environments developed on the MuJoCo physics engine with the Franka Emika Panda arm in MuJoCo Menagerie. Three representative tasks, push, slide, and pick-and-place, are implemented through the Gymnasium Robotics API, which inherits from the core of Gymnasium. Both the sparse binary and dense rewards are supported, and the observation space contains the keys of desired and achieved goals to follow the Multi-Goal Reinforcement Learning framework. Three different off-policy algorithms are used to validate the simulation attributes to ensure the fidelity of all tasks, and benchmark results are also given. Each environment and task are defined in a clean way, and the main parameters for modifying the environment are preserved to reflect the main difference. The repository, including all environments, is available at https://github.com/zichunxx/panda_mujoco_gym.

Objectives: Simulate the transmission process of COVID-19 in a cruise ship, and then to judge how many infections there will be in the 3711 people in the "Diamond Princess" and analyze measures that could have prevented mass transmission. Methods: Based on the crowd flow model, the virus transmission rule between pedestrians is established, to simulate the spread of the virus caused by the close contact during pedestrians' daily activities on the cruise ship. Measurements and main results: Three types of simulation scenarios are designed, the Basic scenario focus on the process of virus transmission caused by a virus carrier and the effect of the personal protective measure against the virus. The condition that the original virus carriers had disembarked halfway and more and more people strengthen self-protection are considered in the Self-protection scenario, which would comparatively accord with the actual situation of "Diamond princess" cruise. Control scenario are set to simulate the effect of taking recommended or mandatory measures on virus transmission Conclusions: There are 850~1009 persons (with large probability) who have been infected with COVID-19 during the voyage of "Diamond Princess". The crowd infection percentage would be controlled effectively if the recommended or mandatory measures can be taken immediately during the alert phase of COVID-19 outbreaks.