In this paper, we proposed a pressure sensor using a Sagnac interferometer based on a side-hole fiber (SHF) with the assistance of deep learning. A convolutional neural network (CNN) was built to identify the spectra of different pressures since the traditional tracing method will face spectral overlap problems when the shift of the spectrum exceeds the free spectral range (FSR). The spectra of pressures ranging from 0 Mpa to 5 MPa with a step of 0.1 MPa will be normalized firstly and then sent to the CNN model for training. The precited result shows that the coefficient of determination$R^{2}$is 99.99987% with the root mean square error (RMSE) equal to$1.6537\times 10^{-3}$MPa. Additionally, a similar structure was constructed to demonstrate the universality of the proposed CNN model. The model can also get a good performance, although the receiving device has a low resolution, showing its great potential for developing a low-cost sensing system.

In this paper, we proposed a laparoscopic grasper force sensor that is able to distinguish both the magnitude and direction of the force, which is improved by random forest. The sensor is based on a seven-core multicore fiber inscribed with Bragg gratings (FBGs) in each core and packaged with polydimethylsiloxane (PDMS). The force measurement was demonstrated by applying the force magnitude in the range of 0 to 6N and direction ranging from 0° to 45°. To realize the prediction of the force and direction simultaneously, a random forest model is established and the results show a magnitude error of 0.104 N and an angular error of 2.269°, implying a good performance even with limited experimental data.

Force redundant mechanism is adopted to achieve high acceleration. However, utilizing the good velocity control performance of servo motor to achieve a high positional accuracy will arise excessive big internal forces. Therefore, an idea of introducing kinematical redundancy into force redundant parallel mechanism is proposed. In this idea, we do not add kinematical redundancy on the sub arm but introduce internal redundant motion into the top plate. The concept of this research is illustrated by improving a simple force redundant parallel example, and the merits of the improved mechanism are summarized. This theory is applied into modifying the real high speed parallel mechanism NINJA. Its kinematics and static force relationship are derived and compared respectively. The prototype of the new mechanism is proposed.

A systematic approach to compute total stiffness of parallel manipulators is proposed. Link stiffness and joint stiffness can be calculated simultaneously. Stiffness of passive joint (joint without actuator) is temporally assumed as nonzero variable in the procedure and infinite stiffness can be treated. A simple parallel manipulator is used as an example to illustrate the process and its correctness. Then validity of the proposed approach is confirmed by comparing to FEA and conventional joint stiffness method. The total link compliance (inverses of total stiffness) calculated by the developed approach and its counterpart obtained by FEA model are very close. The total joint stiffness computed by the developed approach and its correspondence derived from conventional joint stiffness method are totally the same. Stiffness performance of parallel force redundancy mechanisms is evaluated. These results indicate that this scheme is feasible and effective.