Locomotion control has long been vital to legged robots. Agile locomotion can be implemented through either model-based controller or reinforcement learning. It is proven that robust controllers can be obtained through model-based methods and learning-based policies have advantages in generalization. This paper proposed a hybrid framework of locomotion controller that combines deep reinforcement learning and simple heuristic policy and assigns them to different activation phases, which provides guidance for adaptive training without producing conflicts between heuristic knowledge and learned policies. The training in simulation follows a step-by-step stochastic curriculum to guarantee success. Domain randomization during training and assistive extra feedback loops on real robot are also adopted to smooth the transition to the real world. Comparison experiments are carried out on both simulated and real Wukong-IV humanoid robots, and the proposed hybrid approach matches the canonical end-to-end approaches with higher rate of success, faster converging speed, and 60% less tracking error in velocity tracking tasks.

Electric vertical takeoff and landing aircraft (eVTOL) has drawn more and more attention from home and abroad in recent years. It is believed autonomous eVTOL will create a new era of Urban Air Mobility (UAM) and Advanced Air Mobility (AAM). To autonomous eVTOL, precision landing is an extremely critical operation for it directly affects flight safety. In this paper, we analyze the special issues an eVTOL will encounter when it lands in the UAM applications. A multi-stage precision landing method based on multi-marker joint localization is proposed accordingly. Our method contains three key elements: a compatible vertiport with multiple visual markers; an accurate, fast detection and localization algorithm for the vertiport and a multi-stage landing strategy. We implement our method on an eVTOL prototype named ZJ-Copter developed by Zhejiang Lab. A series of real-world experiments have been conducted to validate the effectiveness and accuracy of the proposed method. Experiment results show that our method works well in real-world scenarios for autonomous eVTOL without GPS signal during landing process. The positioning accuracy is less than 0.1m (altitude $< \mathbf{10}\mathbf{m}$ ), while the landing accuracy is less than 0.5m.

Urban air transportation (private aircraft, air taxi, etc.) and all kinds of special flight missions (aerial sightseeing, logistics transportation, disaster relief, anti-terrorism and anti-riot etc.) put forward high requirements for the flight performance of future urban aircraft. It not only demands high-efficiency hovering performance, but also demands the capability of high-speed cruising as well as high safety, clean energy and other characteristics. The maximum flight speed of conventional rotorcraft (such as helicopter, multi-rotor UAV, etc.) is limited by the fluid compressibility of advancing rotor blade and the airflow separation of retreating rotor blade, which makes further improvement of the flight speed more difficult. In order to overcome the fluid compressibility of rotor advancing side and stall of rotor retreating side encountered under high-speed, researchers have proposed many new concepts of advanced rotorcraft. This article sorts and analyses the current advanced VTOL (Vertical Take-Off and Landing) rotorcrafts with new configurations, and attempts to point out the feasible research direction in the future.

Despite their competitive performance on knowledge-intensive tasks, large language models (LLMs) still have limitations in memorizing all world knowledge especially long tail knowledge. In this paper, we study the KG-augmented language model approach for solving the knowledge graph question answering (KGQA) task that requires rich world knowledge. Existing work has shown that retrieving KG knowledge to enhance LLMs prompting can significantly improve LLMs performance in KGQA. However, their approaches lack a well-formed verbalization of KG knowledge, i.e., they ignore the gap between KG representations and textual representations. To this end, we propose an answer-sensitive KG-to-Text approach that can transform KG knowledge into well-textualized statements most informative for KGQA. Based on this approach, we propose a KG-to-Text enhanced LLMs framework for solving the KGQA task. Experiments on several KGQA benchmarks show that the proposed KG-to-Text augmented LLMs approach outperforms previous KG-augmented LLMs approaches regarding answer accuracy and usefulness of knowledge statements.

Advancements in aerial mobility (AAM) are driven by needs in transportation, logistics, rescue, and disaster relief. Consequently, large-sized multirotor unmanned aerial vehicles (UAVs) with strong power and ample space show great potential. In order to optimize the design process for large-sized multirotors and reduce physical trial and error, a detailed dynamic model is firstly established, with an accurate aerodynamic model. In addition, the center of gravity (CoG) offset and actuator dynamics are also well considered, which are usually ignored in small-sized multirotors. To improve the endurance and maneuverability of large-sized multirotors, which is the key concern in real applications, a two-loop optimization method for rotor tilt angle design is proposed based on the mathematical model established previously. Its inner loop solves the dynamic equilibrium points to relax the complex dynamic constraints caused by aerodynamics in the overall optimization problem, which improves the solution efficiency. The ideal design results can be obtained through the offline process, which greatly reduces the difficulties of physical trial and error. Finally, various experiments are carried out to demonstrate the accuracy of the established model and the effectiveness of the optimization method.