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.

The brain–computer interface (BCI) provides direct communication between human brains and machines, including robots, drones and wheelchairs, without the involvement of peripheral systems. BCI based on electroencephalography (EEG) has been applied in many fields, including aiding people with physical disabilities, rehabilitation, education and entertainment. Among the different EEG-based BCI paradigms, steady-state visual evoked potential (SSVEP)-based BCIs are known for their lower training requirements, high classification accuracy and high information transfer rate (ITR). In this article, a filter bank complex spectrum convolutional neural network (FB-CCNN) was proposed, and it achieved leading classification accuracies of 94.85 ± 6.18% and 80.58 ± 14.43%, respectively, on two open SSVEP datasets. An optimization algorithm named artificial gradient descent (AGD) was also proposed to generate and optimize the hyperparameters of the FB-CCNN. AGD also revealed correlations between different hyperparameters and their corresponding performances. It was experimentally demonstrated that FB-CCNN performed better when the hyperparameters were fixed values rather than channel number-based. In conclusion, a deep learning model named FB-CCNN and a hyperparameter-optimizing algorithm named AGD were proposed and demonstrated to be effective in classifying SSVEP through experiments. The hyperparameter design process and analysis were carried out using AGD, and advice on choosing hyperparameters for deep learning models in classifying SSVEP was provided.

Robot hand-eye calibration is crucial for a robot to transform the target 3D position from the camera coordinate system into the robot coordinate system. The conventional method is to use a single-lens camera and calibration board and then solve the AX=XB problem. Recently with the popularization of stereo camera in robot industry, we design a fast and straightforward way to accomplish hand-eye calibration using the point cloud information from the stereo camera and orthogonal projection. We also investigate the number of pairs of data required for calculation in this procedure. We conclude that normally a minimal of 6 points are required for obtaining satisfying result and more pairs of data can provide better result. The procedure is demonstrated and tested on an electric robot arm, and further validated on a full ocean depth hydraulic robot arm. The procedure takes little time to set up and the formulas are concise. It can be applied easily to most robot arms for hand-eye calibration.

Brain-computer interface (BCI) takes commands directly from the user’s brain via electroencephalogram (EEG), which offers a solution for controlling the robotic system when users’ hands are disabled or occupied. This paper describes a semi-autonomous robotic system that is able to execute grasping tasks based on BCI, thus offering an alternative control approach for robotic system. As BCI has limited information transfer rate and can easily cause fatigue in using, we utilize a stereo camera for object localization and convolutional neural network (you only look once version3, YOLOv3) for object identification to reduce the information needed from the brain in task execution, thus alleviating the workload for the user as well as increasing the efficiency of the system.