The texture recognition can provide clues for robots to interact with the external environment. The traditional tactile material recognition task is studied under the close-set assumption, which means that all types of materials are included in the training set. However, the open-set materials recognition for robots is of much greater significance because in the real-world applications, there is usually something that doesn't belong to any known class. Up to now, there is no researcher to further the discussion of this problem. To cope with unknown classes, this study proposes the Open set Material Recognition (OpenMR) based on General Convolutional Prototype Learning (GCPL). To handle the open space risk for GCPL caused by the lack of unknown samples in the training stage, we use Generative Adversarial Networks (GAN) to synthesize open-set samples as unknowns. The proposed framework is implemented and tested on two batches of tactile data collected in different exploratory motions on 8 material textures using the electronic skin. Compared with other open-set classifiers, experiments reveal that the proposed framework achieves competitive performance in both known classification and unknown detection.

Robotic cell injection is a technique that employs automated device to insert substances into a single living cell with a fine needle. Most existing microinjection methods are based on position control without explicit regulation of the injection force. The injection force, if not controlled properly, may destroy the cell and lead to death of the cells. In this paper, we propose a force control based cell injection technology to explicitly regulate the injection force to follow the desired force trajectory during the injection. Any desired force trajectory that is twice continuously differentiable can be realized by the proposed approach. The convergence of the force tracking algorithm is provably guaranteed. Experiments performed on a laboratorial robotic cell injection system demonstrate the effectiveness of the proposed approach.

In a robotic cell injection system, the penetration force applied on the cell reflects the changes of the physical behavior of the cell. The force, if not controlled properly, may damage to the cells or even lead to death of the cells. The current cellular force measurement is limited by the inherent cantilever structure of the sensor, which may not be applicable to a practical cell injection system. In this paper, a simply supported beam structure based PVDF force sensor is first presented. The proportion relation is established between the penetration force and the sensor output after compensation. Using the designed force sensor, the force applied on the cell can be measured, and a force control based cell injection system is constructed. The experimental results performed on zebrafish embryos demonstrate the effectiveness of the micro force sensor and the force based control framework.

Robotic cell microinjection is a technique that employs an automatic method to insert substances into a single living cell with a fine needle. Most available microinjection methods are based on position/velocity tracking, which make it incapable of controlling the injection force. The uncontrolled injection force, however, may destroy the cell and lead to the death of the cell. In this paper, a new adaptive force tracking algorithm within the impedance control framework is proposed to control the injection force applied on the cell. A target impedance is specified in the inner-loop and a trajectory modifying controller is design in the outer-loop for time varying ramp force tracking. The adaptive technique is also used to compensate for the uncertainty of the cell membranepsilas stiffness, the convergence of the system is provably guaranteed. Experiments performed on a laboratorial robotic cell injection system demonstrate the effectiveness of the approach.