基于深度强化学习的无人机栖落机动控制策略设计

无人机栖落机动飞行是一种无需跑道的降落方法，能够提升无人机在复杂环境下执行任务的适应能力。针对具有高非线性、多约束特性的无人机栖落机动过程，提出了一种基于模仿深度强化学习的控制策略设计方法。首先，建立了固定翼无人机栖落机动的纵向非线性动力学模型，并设计了无人机栖落机动的强化学习环境。其次，针对栖落机动状态动作空间大的特点，为了提高探索效率，通过模仿专家经验的方法对系统进行预训练。然后，以模仿学习得到的权重为基础，采用近端策略优化方法学习构建无人机栖落机动的神经网络控制器。最后，通过仿真验证了上述控制策略设计方法的有效性。     

Two-maneuver indirect contingency return from a low lunar orbit

The problem of contingency return from the low lunar orbit is studied. A novel two-maneuver indirect return strategy is proposed. By effectively using the Earth’s gravity to change the orbital plane of the transfer orbit, the second maneuver in the well-known three-maneuver return strategy can be removed, so the total delta-v is reduced. Compared with the single-maneuver direct return, our strategy has the advantage in that the re-entry epoch for the minimum delta-v cost can be advanced in time, with a minimum delta-v value similar to that of the direct return. The most obvious difference between our strategy and the traditional single- or multiple- maneuver strategies is that the complete transfer orbit is a patch between a two-body conic orbit and a three-body orbit instead of two conic orbits. Our strategy can serve as a useful option for contingency return from a low lunar orbit, especially when the delta-v constraint is stringent for a direct return and the contingency epoch is far away from the return window.     

Orbit determination of BeiDou navigation satellite system maneuvered satellites based on instantaneous velocity pulses parameters

The BeiDou navigation satellite system (BDS) comprises geostationary earth orbit (GEO) satellites as well as inclined geosynchronous orbit (IGSO) and medium earth orbit (MEO) satellites. Owing to their special orbital characteristics, GEO satellites require frequent orbital maneuvers to ensure that they operate in a specific orbital window. The availability of the entire system is affected during the maneuver period because service cannot be provided before the ephemeris is restored. In this study, based on the conventional dynamic orbit determination method for navigation satellites, multiple sets of instantaneous velocity pulses parameters which belong to one of pseudo-stochastic parameters were used to simulate the orbital maneuver process in the orbital maneuver arc and establish the observed and predicted orbits of the maneuvered and non-maneuvered satellites of BeiDou regional navigation satellite system (BDS-2) and BeiDou global navigation satellite system (BDS-3). Finally, the single point positioning (SPP) technology was used to verify the accuracy of the observed and predicted orbits. The orbit determination accuracy of maneuvered satellites can be greatly improved by using the orbit determination method proposed in this paper. The overlapping orbit determination accuracy of maneuvered GEO satellites of BDS-2 and BDS-3 can improve 2–3 orders of magnitude. Among them, the radial orbit determination accuracy of each maneuvered satellite is basically better than 1 m. simultaneously, the combined orbit determination of the maneuvered and non-maneuvered satellites does not have a great impact on the orbit determination accuracy of the non-maneuvered satellites. Compared with the multi GNSS products (indicated by GBM) from the German Research Centre for Geosciences (GFZ), the impact of adding the maneuvered satellites on the orbit determination accuracy of BDS-2 satellites is less than 9 %. Furthermore, the orbital recovery time and the service availability period are significantly improved. When the node of the predicted orbit is traversed approximately 3 h after the maneuver, the accuracy of the predicted orbit of the maneuvered satellite can reach that of the observed orbit. The SPP results for the BDS reached a normal level when the node of the predicted orbit was 2 h after the maneuver.     

Closed-form solution of attitude command generation for spin-to-spin maneuver

In recent earth observing missions, agile satellites enable various imaging modes beyond the traditional along-track strip imaging. However, it requires maneuvering with boundary conditions of considerable angular velocity, i.e., spin-to-spin maneuvering. This paper proposes an attitude command generation method for spin-to-spin maneuvering that can provide feedforward commands for the attitude control loop. A general solution for arbitrary flight time is provided which steers a satellite to the given final attitude and angular velocity at the prescribed time. In addition, an alternative method is proposed that further improves the maneuvering speed, which is applicable to small-angle maneuvering cases. The proposed solutions are both closed-form which are more intuitive and easier to comprehend than numerical solutions. It also has a great advantage in computational efficiency, which could enable its use on-board in real time. Numerical examples demonstrate the performance of the proposed methods in a single maneuvering case as well as in a consecutive maneuvering case integrated with a realistic earth observing scenario.     

Energy efficient reactionless design of multi-arm space robot for a cooperative handshake maneuver

Space robots play a significant role in on-orbit capture, space structure construction, and assembly tasks. Since the robotic arms are attached to a free-floating satellite, the motion of the manipulator in such tasks and the satellite are coupled. Multiple-arm space robots can perform complex cooperative tasks and are superior to single-arm space robots. Current work proposes a reactionless manipulation algorithm for a multi-robotic arm based on the iterative Newton–Euler method for space robots with many task and balance arms. The present work demonstrates two tasks and one balance arm to perform a reactionless handshake maneuver in space. This maneuver is presented in detail for a planar and spatial case. The planar case uses 3 DoF robotic arms, while the spatial case uses 6 DoF robotic arms. In addition, the balance arm has been designed considering the efficient usage of energy satisfying reactionless manipulation concept. The design procedure focuses on minimizing energy used during the motion of the balance arm for a known motion of task arms using a genetic algorithm. Moreover, computational experiments are conducted to validate the use of the genetic algorithm for optimization. The results of proposed reactionless manipulation algorithm have been validated with the results available in the literature for the spatial case that uses a different method. In the future, an energy-efficient balance arm will be designed to handle tumbling objects.