Mobile manipulation has a broad range of applications in robotics. However, it is usually more challenging than fixed-base manipulation due to the complex coordination of a mobile base and a manipulator. Although recent works have demonstrated that deep reinforcement learning is a powerful technique for fixed-base manipulation tasks, most of them are not applicable to mobile manipulation. This paper investigates how to leverage deep reinforcement learning to tackle whole-body mobile manipulation tasks in unstructured environments using only on-board sensors. A novel mobile manipulation system which integrates the state-of-the-art deep reinforcement learning algorithms with visual perception is proposed. It has an efficient framework decoupling visual perception from the deep reinforcement learning control, which enables its generalization from simulation training to real-world testing. Extensive simulation and experiment results show that the proposed mobile manipulation system is able to grasp different types of objects autonomously in various simulation and real-world scenarios, verifying the effectiveness of the proposed mobile manipulation system.

The brain–computer interface (BCI), which provides a new way for humans to directly communicate with robots without the involvement of the peripheral nervous system, has recently attracted much attention. Among all the BCI paradigms, BCIs based on steady-state visual evoked potentials (SSVEPs) have the highest information transfer rate (ITR) and the shortest training time. Meanwhile, deep learning has provided an effective and feasible solution for solving complex classification problems in many fields, and many researchers have started to apply deep learning to classify SSVEP signals. However, the designs of deep learning models vary drastically. There are many hyper-parameters that influence the performance of the model in an unpredictable way. This study surveyed 31 deep learning models (2011–2023) that were used to classify SSVEP signals and analyzed their design aspects including model input, model structure, performance measure, etc. Most of the studies that were surveyed in this paper were published in 2021 and 2022. This survey is an up-to-date design guide for researchers who are interested in using deep learning models to classify SSVEP signals.

Recently, as humans have become increasingly interested in ocean resources, underwater vehicle-manipulator systems (UVMSs) have played an increasingly important role in ocean exploitation. To realize precise operation in underwater narrow spaces, the fly arm underwater vehicle manipulator system (FAUVMS) is proposed with manipulators as its core. However, this system suffers severe dynamic coupling effects due to the combination of small vehicle and big manipulators. To resolve this issue, we propose a robust adaptive controller that contains two parts. In the first part, the extended Kalman filter (EKF) is designed to estimate the system states and predicts external disturbances to achieve adaptive control. In the second part, a chattering-free sliding mode control (SMC) is designed to converge the tracking errors to zero, thus guaranteeing the robustness of the controller. We constructed the simulation platform based on the geometric model of FAUVMS, and various simulations are carried out under different situations. Compared to the traditional methods, the proposed method has a faster convergent speed, a better robustness and adaptiveness to external disturbances, and the tracking errors of positions of the vehicle and each end-effector are much smaller.

This letter proposes a learning-based adaptive propulsion control (APC) method for a quadruped robot integrated with thrusters in amphibious environments, allowing it to move efficiently in water while maintaining its ground locomotion capabilities. We designed the specific reinforcement learning method to train the neural network to perform the vector propulsion control. Our approach coordinates the legs and propeller, enabling the robot to achieve speed and trajectory tracking tasks in the presence of actuator failures and unknown disturbances. Our simulated validations of the robot in water demonstrate the effectiveness of the trained neural network to predict the disturbances and actuator failures based on historical information, showing that the framework is adaptable to changing environments and is suitable for use in dynamically changing situations. Our proposed approach is suited to the hardware augmentation of quadruped robots to create avenues in the field of amphibious robotics and expand the use of quadruped robots in various applications.

The dynamic performance and load capacity of a deep-sea electric manipulator are obviously affected by the viscous resistance of compensating oil. However, the complex geometric clearance of the oil-filled joint of the manipulator makes it difficult for existing theoretical models to accurately model this problem. Based on the existing theoretical models, a viscous power model considering the influence of the surface grooves of the rotor and stator is proposed. To reduce the viscous power loss, an optimization method for the clearance geometry is proposed. This optimization method is carried out by filling the surface grooves of the motor rotor and stator with epoxy resin. This method makes the clearance geometry of the compensated oil more regular. The modified model and optimization method are validated by designing an experimental installation and method. The experimental results demonstrate the accuracy and effectiveness of the modified theoretical model and optimization method.