Population growth, climate change, and the worldwide COVID-19 pandemic are imposing increasing pressure on global agricultural production. The challenge of increasing crop yield while ensuring sustainable development of environmentally friendly agriculture is a common issue throughout the world. Autonomous systems, sensing technologies, and artificial intelligence offer great opportunities to tackle this issue. In precision agriculture (PA), non-destructive and non-invasive remote and proximal sensing methods have been widely used to observe crops in visible and invisible spectra. Nowadays, the integration of high-performance imagery sensors (e.g., RGB, multispectral, hyperspectral, thermal, and SAR) and unmanned mobile platforms (e.g., satellites, UAVs, and terrestrial agricultural robots) are yielding a huge number of high-resolution farmland images, in which rich crop information is compressed. However, this has been accompanied by challenges, i.e., ways to swiftly and efficiently making full use of these images, and then, to perform fine crop management based on information-supported decision making. In the past few years, deep learning (DL) has shown great potential to reshape many industries because of its powerful capabilities of feature learning from massive datasets, and the agriculture industry is no exception. More and more agricultural scientists are paying attention to applications of deep learning in image-based farmland observations, such as land mapping, crop classification, biotic/abiotic stress monitoring, and yield prediction. To provide an update on these studies, we conducted a comprehensive investigation with a special emphasis on deep learning in multiscale agricultural remote and proximal sensing. Specifically, the applications of convolutional neural network-based supervised learning (CNN-SL), transfer learning (TL), and few-shot learning (FSL) in crop sensing at land, field, canopy, and leaf scales are the focus of this review. We hope that this work can act as a reference for the global agricultural community regarding DL in PA and can inspire deeper and broader research to promote the evolution of modern agriculture.

The loaded walking ability of people can be enhanced by lower limb exoskeleton. How to avoid lower limb joint injuries while improving walking efficiency is a major challenge for loaded walking assistance. Inspired by the muscle-skeleton working mechanism of human, we proposed a lower limb exoskeleton with rigid and soft structure for active hip extension assistance and passive load transfer assistance. Its motion flexibility, load transfer ability and metabolic cost were investigated. Five motions including jumping jack, sit down, crouch, across obstacle, and forward lunge were realized by the wearer. Under the conditions of quiet standing and dynamic walking, the percentages indicating the degree of load transfer to the ground reached up to 88% and 67%, respectively. Compared to the condition of no exoskeleton, the net metabolic reductions of 20kg-loaded walking in three experimental setups, namely, active-passive assistance on, active exosuit assistance only, and active assistance off-passive assistance on, achieved 12.8%, 8.5% and 3.7%, individually. The results demonstrate that our rigid-soft lower limb exoskeleton presents outperformance in loaded walking assistance.

In precision agriculture, unmanned aerial vehicles (UAVs) are playing an increasingly important role in farmland information acquisition and fine management. However, discrete obstacles in the farmland environment, such as trees and power lines, pose serious threats to the flight safety of UAVs. Real-time detection of the attributes of obstacles is urgently needed to ensure their flight safety. In the wake of rapid development of deep learning, object detection algorithms based on convolutional neural networks (CNN) and transformer architectures have achieved remarkable results. Detection Transformer (DETR) and Deformable DETR combine CNN and transformer to achieve end-to-end object detection. The goal of this work is to use Deformable DETR for the task of farmland obstacle detection from the perspective of UAVs. However, limited by local receptive fields and local self-attention mechanisms, Deformable DETR lacks the ability to capture long-range dependencies to some extent. Inspired by non-local neural networks, we introduce the global modeling capability to the front-end ResNet to further improve the overall performance of Deformable DETR. We refer to the improved version as Non-local Deformable DETR. We evaluate the performance of Non-local Deformable DETR for farmland obstacle detection through comparative experiments on our proposed dataset. The results show that, compared with the original Deformable DETR network, the mAP value of the Non-local Deformable DETR is increased from 71.3% to 78.0%. Additionally, Non-local Deformable DETR also presents great performance for detecting small and slender objects. We hope this work can provide a solution to the flight safety problems encountered by UAVs in unstructured farmland environments.

With the development of precision agriculture (PA), low-altitude and low-volume spraying based on unmanned aerial vehicles (UAVs) is playing an increasingly important role in the control of crop diseases, pests, and weeds. However, the aerial spraying quality and droplet drift are affected by many factors, some of which are controllable (e.g., flight and spraying parameters) and some of which are not (e.g., environmental parameters). In order to study the influence of spraying parameters on the UAV-based spraying performance, we propose a UAV-compatible spraying system and a customized experimental platform in this work. Through single-factor test and Box–Behnken response surface methods, four influencing factors, namely spraying height, flow rate, distance between nozzles, and pulse width modulation (PWM) duty cycle, were studied under indoor conditions. Variance analysis and multiple quadratic regression fitting were performed on the test data by using Design-Expert 8.0.5B software, and quadratic polynomial regression models of effective spraying width, droplet coverage density, coefficient of variation, and droplet coverage rate were established. Based on the Z-score standardization, a mathematical model of the comprehensive score with four factors was established to evaluate the spraying quality and predict optimal spraying parameters. Test results indicate that the effect intensity of four influencing factors from strong to weak is PWM duty cycle, flow rate, distance between nozzles, and spraying height, and their optimal values are 98.65%, 1.74 L/min, 1.0 m, and 1.60 m, respectively. Additionally, verification experimental results demonstrate that the deviation between the predicted comprehensive score and the actual value was less than 6%. This paper can provide a reference for the design and optimization of UAV spraying systems.