Unmanned aerial systems (UAS) are increasingly used in precision agriculture to collect crop health related data. UAS can capture data more often and more cost-effectively than sending human scouts into the field. However, in large crop fields, flight time, and hence data collection, is limited by battery life. In a conventional UAS approach, human operators are required to exchange depleted batteries many times, which can be costly and time consuming. In this study, we developed a novel, fully autonomous aerial scouting approach that preserves battery life by sampling sections of a field for sensing and predicting crop health for the whole field. Our approach uses reinforcement learning (RL) and convolutional neural networks (CNN) to accurately and autonomously sample the field. To develop and test the approach, we ran flight simulations on an aerial image dataset collected from an 80-acre corn field. The excess green vegetation Index was used as a proxy for crop health condition. Compared to the conventional UAS scouting approach, the proposed scouting approach sampled 40% of the field, predicted crop health with 89.8% accuracy, reduced labor cost by 4.8× and increased agricultural profits by 1.36×.

A critical review is provided to summarize our recent efforts to utilize the state-of-the-art bio-inspired icephobic coatings/surfaces, i.e., 1). Lotus-leaf-inspired superhydrophobic surfaces (SHS) and 2). Pitcher-plant-inspired slippery liquid-infused porous surfaces (SLIPS) for aircraft icing mitigation. By leveraging the unique Icing Research Tunnel of Iowa State University (i.e., ISU-IRT), an experimental campaign was performed to evaluate the effectiveness of using SHS and SLIPS coatings to suppress impact ice accretion over the surfaces of typical airfoil/wing models. While both SHS and SLIPS were found to be able to suppress ice accretion over the airframe surfaces where strong aerodynamic forces are exerted, ice was still found to accrete in the vicinity of the airfoil stagnation line where the aerodynamic forces are at their minimum. A novel hybrid anti-/de-icing strategy concept to combine icephobic coatings with minimized surface heating near airfoil leading edge was demonstrated to effectively remove impact ice accretion over entire airfoil/wing surfaces. An experimental investigation was also conducted to examine the durability of the icephobic coatings/surfaces to resist "rain erosion" effects (i.e., the damage to the surface coatings due to continuous impingement of water droplets at high speeds) in considering the practical usage for aircraft icing mitigation. The rain erosion effects were characterized based on the variations of the ice adhesion strengths and surface morphology of the eroded test surfaces coated with SHS and SLIPS. The research findings are very helpful to elucidate the underlying physics for the development of novel and robust anti-/de-icing strategies for aircraft icing mitigation.

We propose an end-to-end neural network that improves the segmentation accuracy of fully convolutional networks by incorporating a localization unit. This network performs object localization first, which is then used as a cue to guide the training of the segmentation network. We test the proposed method on a segmentation task of small objects on a clinical dataset of ultrasound images. We show that by jointly learning for detection and segmentation, the proposed network is able to improve the segmentation accuracy compared to only learning for segmentation.

Aircraft icing is widely recognized as one of the most serious weather hazards to flight safety. Specially designed hydro-/ice-phobic coatings are currently undergoing development for aircraft icing mitigation. It was found that hydro-/icephobic coatings would delay the ice accretion iover airframe surfaces so that the impacted supercooled water droplets could be blown away by the airflow from the airframe surface before being frozen into ice. It is of fundamental importance to understand the wind-driven runback behavior of water droplets over surfaces treated with different coatings, since the corresponding knowledge would be very helpful and essential to develop more efficient anti-/de-icing systems for aircraft icing protection. With the rapid development of surface engineering, a series of specially designed surface coatings succeed in icing mitigation using airflow to remove the remained water. While various hydro-/ice-phobic coatings/surfaces have been developed in recent years, the “state-of-the-art” icephobic coatings/surfaces can be generally divided into three categories, i.e., 1). Lotus-leaf-inspired superhydrophobic surfaces (SHS) with micro-/nano-scale surface textures to achieve very high contact angles (typically > 150°); 2). Pitcherplant-inspired slippery liquid infused porous surfaces (SLIPS) with a layer of liquid lubricant (which is immiscible with water) being sandwiched between ice and solid substrate materials; and 3). Icephobic elastic materials/surfaces with deformable structures/surfaces. SHS has a water droplet contact angle (CA) larger than 150° and a sliding angle (SA) less than 10° . SHS always has a hierarchical structure which is similar to the lotus leaf, and water droplets on SHS appear as water beads which can easily roll off the surface by wind or gravity before frozen. Another strategy to reduce ice adhesion strength to a solid surface is to use a layer of liquid lubricant, which is immiscible with water, between ice and the solid surface. The use of such lubricated surfaces was investigated as early as 1960s, and has gained increasing attentions again recently with the introduction of a concept called Slippery Liquid-Infused Porous Surfaces (SLIPS). SLIPS concept is inspired by the Nepenthes pitcher plants, which have evolved highly slippery, liquidinfused micro-textured rim to capture insects. SLIPS surfaces were not only found to be able to suppress ice/frost accretion by effectively removing condensed moisture even in high humidity conditions, but also exhibit at least an order of magnitude lower ice adhesion than most SHS coatings. More recently, elastic materials/surfaces, such as Polydimethylsiloxane or PDMS in short, which would be structurally deformed/altered dynamically upon applying extra mechanical stress, have also been suggested for icing mitigation. Elastic materials display ultra-low adhesion to ice due to their low work of adhesion and liquidlike deformability, while maintaining good mechanical durability due to their solid-like rigidity. It is found that water droplets would not only be more readily rebounding away from the surface after impingement, but also be able to roll away before frozen due to the hydrophobicity of PDMS. Considering the differences in wettabilities and mechanisms of water repellency, it is necessary to have a systematic understanding of how efficient the surfaces are when the aerodynamic force is applied to remove the adhered water droplets. In the present study, a comprehensive experimental campaign was conducted to characterize the transient runback behaviors of wind-driven water droplets over the surfaces of test plates coated with different hydro-/icephobic coatings (i.e., SHS, SLIPS and PDMS). A high-resolution Particle Image Velocimetry (PIV) system was used to achieve quantitative measurements of the velocity field of the airflow around the wind-driven water droplets on the test surfaces with different wettabilities. With the detailed PIV measurements of the airflow field around the runback water droplets and the droplet profiles, the aerodynamic forces and the adhesion forces acting on the water droplets were estimated. While Fig. 1 shows the experimental setup used in the present study, Fig. 2 to Fig. 3 given some of the typical measurement results. More measurement results and comprehensive analysis and discussions will be provided in the full version of this research paper.