Artificial Intelligence (AI) and automation are fostering more sustainable and effective solutions for a wide spectrum of agricultural problems. Pest management is a major challenge for crop production that can benefit from machine learning techniques to detect and monitor specific pests and diseases. Traditional monitoring is labor intensive, time demanding, and expensive, while machine learning paradigms may support cost-effective crop protection decisions. However, previous studies mainly relied on morphological images of stationary or immobilized animals. Other features related to living animals behaving in the environment (e.g., walking trajectories, different postures, etc.) have been overlooked so far. In this study, we developed a detection method based on convolutional neural network (CNN) that can accurately classify in real-time two tephritid species (Ceratitis capitata and Bactrocera oleae) free to move and change their posture. Results showed a successful automatic detection (i.e., precision rate about 93%) in real-time of C. capitata and B. oleae adults using a camera sensor at a fixed height. In addition, the similar shape and movement patterns of the two insects did not interfere with the network precision. The proposed method can be extended to other pest species, needing minimal data pre-processing and similar architecture.

Motor performance deficits are major issues impacting functional activities in older adults. Herein, we used locusts, emergent model organisms for neurodegeneration and motor deficit studies, to investigate the role of cognition as compensation strategy to age-related motor activation deficiency. An animal-robot interaction has been established to evoke anti-predator motor responses in locusts. We found a significantly earlier decision making of older adults that displayed their first motor response at a significantly longer distance from the approaching robotic predator, compared to young adults. Although older adult locusts reacted earlier to the approaching robot, performing a wider movement with their hind leg if compared to young adults, older adults obtained a body displacement significantly shorter than the body displacement of younger individuals. The earlier reaction in older adult locusts is probably due to a strategic behavior compensating a slower muscular response that may require a longer time to be performed, and this is consistent with other compensation strategies reported in humans. To the best of our knowledge this research represents a pilot study using ethorobotics to characterize associations between cognitive and motor functions, adding basic information on high-speed sensory-motor activation, that can be exploited to ameliorate motor decline in older adults.

Torque and force signals data were acquired from a load-cell sensor during a robotic welding process, in presence of collisions between the tool and the workpiece edges outlined in part in "Haptic-based touch detection for collaborative robots in welding applications" [1]. The dataset is composed from 15 tests captured during a tele-operated welding robot performing a 1G ASME/AWS (i.e., PA ISO) welding process. The raw data files have been provided. These data can be used to correlate torque signal features with collision events, to improve algorithms of collision detection/avoidance and to develop reliable real-time haptic feedback to the welder. This dataset can also be used to study the torque signal variation in different welding positions (e.g., 2G, 3G, 2F, etc.). Dataset is provided as raw data and in MATLAB files.