Work-related musculoskeletal disorder (WMSD) is a common occupational injury. Among the occupational injuries of agricultural workers, 93% are related to WMSD, with the shoulder being the second most affected area. This paper presents a passive upper-limb exoskeleton (PULE) with a gas spring and four-bar mechanism developed to aid the daily activities associated with orchard farming. The PULE is used to assist the arm-lifting process, reducing the physical exertion of farmers and the risk of developing WMSD. Electromyography (EMG) measurements of 26 participants were obtained to evaluate the difference in physical exertion with and without the PULE. Two arm activities: fruit thinning, with the arms raised and maintained at 100°, and pesticide spraying, with the hands swinging from 0° to 100° and back, were simulated. Using the PULE decreased muscle tension of the anterior deltoid (AD) by 17.64–19.86%. The PULE also decreased the AD activity by 37.67–39.57% during the actual orchard farming operations. The Qualisys motion capture system indicated that the difference in the lifting angle of the upper limb with and without the PULE was less than 1° and not significant (α > 0.05). Thus, the PULE did not affect the flexibility of the wearer in orchard farming activities.

Destemming fresh chilli fruit (Capsicum) in large productivity is necessary, especially in the Mekong Delta region. Several studies have been done to solve this problem with high applicability, but a certain percentage of the output consisted of cracked fruits, thus reducing the quality of the system. The manual sorting results in high costs and low quality, so it is necessary that automatic grading is performed after destemming. This research focused on developing a method to identify and classify cracked chilli fruits caused by the destemming process. The convolution neural network (CNN) model was built and trained to identify cracks; then, appropriate control signals were sent to the actuator for classification. Image processing operations are supported by the OpenCV library, while the TensorFlow data structure is used as a database and the Keras application programming interface supports the construction and training of neural network models. Experiments were carried out in both the static and working conditions, which, respectively, achieved an accurate identification rate of 97 and 95.3%. In addition, a success rate of 93% was found even when the chilli body is wrinkled due to drying after storage time at 120 hours. Practical results demonstrate that the reliability of the model was useful and acceptable.

There are the diseased fruits left in the product of the automatic fresh chili destemming system. This will reduce the processing quality, so it is necessary to develop a grading system. This research focused on building a model to identify the damages caused by diseases on fresh destemmed chili fruits. A convolution neural network (CNN) model was built and trained with the support of OpenCV, TensorFlow, and Keras libraries in order to identify damages on the fruit body. An independent test was performed on 184 fruits with 1840 identify times and achieved an accuracy rate of 90.8%. This result was acceptable and would be used as a basis for the future classification system. Further improvements could be investigated to achieve a higher success rate.