The detection of cherry tomatoes in greenhouse scene is of great significance for robotic harvesting. This paper states a method based on deep learning for cherry tomatoes detection to reduce the influence of illumination, growth difference, and occlusion. In view of such greenhouse operating environment and accuracy of deep learning, Single Shot multi-box Detector (SSD) was selected because of its excellent anti-interference ability and self-taught from datasets. The first step is to build datasets containing various conditions in greenhouse. According to the characteristics of cherry tomatoes, the image samples with illumination change, images rotation and noise enhancement were used to expand the datasets. Then training datasets were used to train and construct network model. To study the effect of base network and the input size of networks, one contrast experiment was designed on different base networks of VGG16, MobileNet, Inception V2 networks, and the other contrast experiment was conducted on changing the network input image size of 300 pixels by 300 pixels, 512 pixels by 512 pixels. Through the analysis of the experimental results, it is found that the Inception V2 network is the best base network with the average precision of 98.85% in greenhouse environment. Compared with other detection methods, this method shows substantial improvement in cherry tomatoes detection.

Effective propulsion of untethered micro-/nanorobots at low Reynolds numbers can offer possibilities for promising biomedical applications. Diverse locomotion modes have been proposed for propulsion at a small scale, and rolling is an alternative method which is significantly effective. Here, we demonstrate mass produced magnetic wheel-shape flaky microswimmers fabricated via a simple and cost-effective method. Locomotion behaviors under vertical rotating magnetic fields were studied, and the propulsion mechanisms were analyzed. They exhibited two modes to swim forward as tumbling and rolling, which relied on the actuating field and the fluid. The rolling microswimmers could be propelled and steered precisely and a high velocity can be easily reached. Forward velocity and transition frequency within diverse fields and fluids were analyzed, and side slip effects when rolling at a camber angle were also observed. Such microswimmers synthesized in bulk with alternative locomotion modes and excellent swimming performances may have potential in low Reynolds number fluids.

Marine hydrological information has a significant impact on human development and the utilization of the oceans, which can be monitored with underwater sensors. In the past, ocean research has relied on the use of bulky underwater recorders and sensory telemetry networks. In this study, an integrated flexible sensor is developed for underwater conductivity, temperature, depth, and strain detection. Platinum resistance sensors were used for temperature and strain measurements, conductivity sensors with interdigitated electrodes were used for salinity measurements, and capacitive pressure sensors for depth measurements. Two kinds of flexible capacitive pressure sensors were fabricated with a carbon nanotube/polydimethylsiloxane (CNT/PDMS) bottom electrode and copper/polyimide (Cu/PI) bottom electrode. The sensor with the CNT/PDMS bottom electrode outperformed the sensor with the Cu/PI bottom electrode over a wide pressure range (<5 MPa) and showed stable capacitance up to 1000 cycles. COMSOL simulations also support our experimental results with high sensitivity of the sensor with a CNT/PDMS bottom electrode. The integrated flexible sensor is durable and lightweight, making it ideal for use as a stationary monitoring sensor or for attachment to a variety of marine animals.