Search for: All records

Abstract We present a position and orientation controller for a hybrid rigid-soft manipulator arm where the soft arm is extruded from a two degrees-of-freedom rigid link. Our approach involves learning the dynamics of the hybrid arm operating at 4Hz and leveraging it to generate optimal trajectories that serve as expert data to learn a control policy. We performed an extensive evaluation of the policy on a physical hybrid arm capable of jointly controlling rigid and soft actuation. We show that with a single policy, the arm is capable of reaching arbitrary poses in the workspace with 3.73cm (<6% overall arm length) and 17.78 deg error within 12.5s, operating at different control frequencies, and controlling the end effector with different loads. Our results showcase significant improvements in control speed while effectively controlling both the position and orientation of the end effector compared to previous quasistatic controllers for hybrid arms. 

IntroductionEffective monitoring of insect-pests is vital for safeguarding agricultural yields and ensuring food security. Recent advances in computer vision and machine learning have opened up significant possibilities of automated persistent monitoring of insect-pests through reliable detection and counting of insects in setups such as yellow sticky traps. However, this task is fraught with complexities, encompassing challenges such as, laborious dataset annotation, recognizing small insect-pests in low-resolution or distant images, and the intricate variations across insect-pests life stages and species classes. MethodsTo tackle these obstacles, this work investigates combining two solutions, Hierarchical Transfer Learning (HTL) and Slicing-Aided Hyper Inference (SAHI), along with applying a detection model. HTL pioneers a multi-step knowledge transfer paradigm, harnessing intermediary in-domain datasets to facilitate model adaptation. Moreover, slicing-aided hyper inference subdivides images into overlapping patches, conducting independent object detection on each patch before merging outcomes for precise, comprehensive results. ResultsThe outcomes underscore the substantial improvement achievable in detection results by integrating a diverse and expansive in-domain dataset within the HTL method, complemented by the utilization of SAHI. DiscussionWe also present a hardware and software infrastructure for deploying such models for real-life applications. Our results can assist researchers and practitioners looking for solutions for insect-pest detection and quantification on yellow sticky traps. 

Abstract The recent development and use of generative AI (GenAI) has signaled a significant shift in research activities such as brainstorming, proposal writing, dissemination, and even reviewing. This has raised questions about how to balance the seemingly productive uses of GenAI with ethical concerns such as authorship and copyright issues, use of biased training data, lack of transparency, and impact on user privacy. To address these concerns, many Higher Education Institutions (HEIs) have released institutional guidance for researchers. To better understand the guidance that is being provided we report findings from a thematic analysis of guidelines from thirty HEIs in the United States that are classified as R1 or “very high research activity.” We found that guidance provided to researchers: (1) asks them to refer to external sources of information such as funding agencies and publishers to keep updated and use institutional resources for training and education; (2) asks them to understand and learn about specific GenAI attributes that shape research such as predictive modeling, knowledge cutoff date, data provenance, and model limitations, and educate themselves about ethical concerns such as authorship, attribution, privacy, and intellectual property issues; and (3) includes instructions on how to acknowledge sources and disclose the use of GenAI, how to communicate effectively about their GenAI use, and alerts researchers to long term implications such as over reliance on GenAI, legal consequences, and risks to their institutions from GenAI use. Overall, guidance places the onus of compliance on individual researchers making them accountable for any lapses, thereby increasing their responsibility. 

Abstract Data augmentation is a powerful tool for improving deep learning‐based image classifiers for plant stress identification and classification. However, selecting an effective set of augmentations from a large pool of candidates remains a key challenge, particularly in imbalanced and confounding datasets. We propose an approach for automated class‐specific data augmentation using a genetic algorithm. We demonstrate the utility of our approach on soybean [Glycine max (L.) Merr] stress classification where symptoms are observed on leaves; a particularly challenging problem due to confounding classes in the dataset. Our approach yields substantial performance, achieving a mean‐per‐class accuracy of 97.61% and an overall accuracy of 98% on the soybean leaf stress dataset. Our method significantly improves the accuracy of the most challenging classes, with notable enhancements from 83.01% to 88.89% and from 85.71% to 94.05%, respectively. A key observation we make in this study is that high‐performing augmentation strategies can be identified in a computationally efficient manner. We fine‐tune only the linear layer of the baseline model with different augmentations, thereby reducing the computational burden associated with training classifiers from scratch for each augmentation policy while achieving exceptional performance. This research represents an advancement in automated data augmentation strategies for plant stress classification, particularly in the context of confounding datasets. Our findings contribute to the growing body of research in tailored augmentation techniques and their potential impact on disease management strategies, crop yields, and global food security. The proposed approach holds the potential to enhance the accuracy and efficiency of deep learning‐based tools for managing plant stresses in agriculture. 

Abstract Spatial adjustments are used to improve the estimate of plot seed yield across crops and geographies. Moving means (MM) and P‐Spline are examples of spatial adjustment methods used in plant breeding trials to deal with field heterogeneity. Within the trial, spatial variability primarily comes from soil feature gradients, such as nutrients, but a study of the importance of various soil factors including nutrients is lacking. We analyzed plant breeding progeny row (PR) and preliminary yield trial (PYT) data of a public soybean breeding program across 3 years consisting of 43,545 plots. We compared several spatial adjustment methods: unadjusted (as a control), MM adjustment, P‐spline adjustment, and a machine learning‐based method called XGBoost. XGBoost modeled soil features at: (a) the local field scale for each generation and per year, and (b) all inclusive field scale spanning all generations and years. We report the usefulness of spatial adjustments at both PR and PYT stages of field testing and additionally provide ways to utilize interpretability insights of soil features in spatial adjustments. Our work shows that using soil features for spatial adjustments increased the relative efficiency by 81%, reduced the similarity of selection by 30%, and reduced the Moran's I from 0.13 to 0.01 on average across all experiments. These results empower breeders to further refine selection criteria to make more accurate selections and select for macro‐ and micro‐nutrients stress tolerance. 

Abstract Insect pests significantly impact global agricultural productivity and crop quality. Effective integrated pest management strategies require the identification of insects, including beneficial and harmful insects. Automated identification of insects under real-world conditions presents several challenges, including the need to handle intraspecies dissimilarity and interspecies similarity, life-cycle stages, camouflage, diverse imaging conditions, and variability in insect orientation. An end-to-end approach for training deep-learning models, InsectNet, is proposed to address these challenges. Our approach has the following key features: (i) uses a large dataset of insect images collected through citizen science along with label-free self-supervised learning to train a global model, (ii) fine-tuning this global model using smaller, expert-verified regional datasets to create a local insect identification model, (iii) which provides high prediction accuracy even for species with small sample sizes, (iv) is designed to enhance model trustworthiness, and (v) democratizes access through streamlined machine learning operations. This global-to-local model strategy offers a more scalable and economically viable solution for implementing advanced insect identification systems across diverse agricultural ecosystems. We report accurate identification (>96% accuracy) of numerous agriculturally and ecologically relevant insect species, including pollinators, parasitoids, predators, and harmful insects. InsectNet provides fine-grained insect species identification, works effectively in challenging backgrounds, and avoids making predictions when uncertain, increasing its utility and trustworthiness. The model and associated workflows are available through a web-based portal accessible through a computer or mobile device. We envision InsectNet to complement existing approaches, and be part of a growing suite of AI technologies for addressing agricultural challenges. 

Abstract Developments in genomics and phenomics have provided valuable tools for use in cultivar development. Genomic prediction (GP) has been used in commercial soybean [Glycine maxL. (Merr.)] breeding programs to predict grain yield and seed composition traits. Phenomic prediction (PP) is a rapidly developing field that holds the potential to be used for the selection of genotypes early in the growing season. The objectives of this study were to compare the performance of GP and PP for predicting soybean seed yield, protein, and oil. We additionally conducted genome‐wide association studies (GWAS) to identify significant single‐nucleotide polymorphisms (SNPs) associated with the traits of interest. The GWAS panel of 292 diverse accessions was grown in six environments in replicated trials. Spectral data were collected at two time points during the growing season. A genomic best linear unbiased prediction (GBLUP) model was trained on 269 accessions, while three separate machine learning (ML) models were trained on vegetation indices (VIs) and canopy traits. We observed that PP had a higher correlation coefficient than GP for seed yield, while GP had higher correlation coefficients for seed protein and oil contents. VIs with high feature importance were used as covariates in a new GBLUP model, and a new random forest model was trained with the inclusion of selected SNPs. These models did not outperform the original GP and PP models. These results show the capability of using ML for in‐season predictions for specific traits in soybean breeding and provide insights on PP and GP inclusions in breeding programs. 

Abstract Soybean (Glycine max[L.] Merr.) production is susceptible to biotic and abiotic stresses, exacerbated by extreme weather events. Water limiting stress, that is, drought, emerges as a significant risk for soybean production, underscoring the need for advancements in stress monitoring for crop breeding and production. This project combined multi‐modal information to identify the most effective and efficient automated methods to study drought response. We investigated a set of diverse soybean accessions using multiple sensors in a time series high‐throughput phenotyping manner to: (1) develop a pipeline for rapid classification of soybean drought stress symptoms, and (2) investigate methods for early detection of drought stress. We utilized high‐throughput time‐series phenotyping using unmanned aerial vehicles and sensors in conjunction with machine learning analytics, which offered a swift and efficient means of phenotyping. The visible bands were most effective in classifying the severity of canopy wilting stress after symptom emergence. Non‐visual bands in the near‐infrared region and short‐wave infrared region contribute to the differentiation of susceptible and tolerant soybean accessions prior to visual symptom development. We report pre‐visual detection of soybean wilting using a combination of different vegetation indices and spectral bands, especially in the red‐edge. These results can contribute to early stress detection methodologies and rapid classification of drought responses for breeding and production applications. 

Abstract BackgroundCOVID‐19 has led to an unprecedented increase in the use of technology for teaching and learning in higher education institutions (HEIs), including in engineering, computing, and technology programs. Given the urgency of the situation, technologies were often implemented with a short‐term rather than long‐term view. PurposeIn this study, we investigate students' perceptions of the use of video‐based monitoring (VbM) for proctoring exams to better assess its impact on students. We leverage technological ambivalence as a framing lens to analyze students' experiences and perceptions of using VbM and draw implications for responsible use of educational technology. MethodQualitative data were collected from students using focus group interviews and discussion board assignments and analyzed inductively to understand students' experiences. FindingsWe present a framework of how a technological shift of existing practice triggered ambivalence that manifested itself as a sustained negative outlook among students regarding the use of VbM, as well as their institution and instructors. Students accepted the inevitability of the technology but were unconvinced that the benefits of VbM outweighed its risks. ConclusionsAs instructors use educational technologies that are inherently driven by user data and algorithms that are not transparent, it is imperative that they are attentive to the responsible use of technology. To educate future engineers who are ethically and morally responsible, engineering educators and engineering institutions need to exhibit that behavior in their own practices, starting with their use of educational technologies. 

IntroductionComputer vision and deep learning (DL) techniques have succeeded in a wide range of diverse fields. Recently, these techniques have been successfully deployed in plant science applications to address food security, productivity, and environmental sustainability problems for a growing global population. However, training these DL models often necessitates the large-scale manual annotation of data which frequently becomes a tedious and time-and-resource- intensive process. Recent advances in self-supervised learning (SSL) methods have proven instrumental in overcoming these obstacles, using purely unlabeled datasets to pre-train DL models. MethodsHere, we implement the popular self-supervised contrastive learning methods of NNCLR Nearest neighbor Contrastive Learning of visual Representations) and SimCLR (Simple framework for Contrastive Learning of visual Representations) for the classification of spatial orientation and segmentation of embryos of maize kernels. Maize kernels are imaged using a commercial high-throughput imaging system. This image data is often used in multiple downstream applications across both production and breeding applications, for instance, sorting for oil content based on segmenting and quantifying the scutellum’s size and for classifying haploid and diploid kernels. Results and discussionWe show that in both classification and segmentation problems, SSL techniques outperform their purely supervised transfer learning-based counterparts and are significantly more annotation efficient. Additionally, we show that a single SSL pre-trained model can be efficiently finetuned for both classification and segmentation, indicating good transferability across multiple downstream applications. Segmentation models with SSL-pretrained backbones produce DICE similarity coefficients of 0.81, higher than the 0.78 and 0.73 of those with ImageNet-pretrained and randomly initialized backbones, respectively. We observe that finetuning classification and segmentation models on as little as 1% annotation produces competitive results. These results show SSL provides a meaningful step forward in data efficiency with agricultural deep learning and computer vision.