To address large-scale problems such as ocean cleanup and reforestation, engineers need to develop novel solutions that can effectively cover a wide area and adapt to a variety of situations. To design a complex system that can solve these problems, such as a fleet of devices, it is necessary to have methods that can guide engineers towards optimal designs. However, determining the optimal engineering specifications is a difficult task because there is a complex relationship between the individual device and how the system performs as a whole. This work develops a method to find these specifications using an AI-guided design method based on human-in-the-loop Bayesian optimization. By representing design tradeoffs using machine learning, the method can effectively guide engineers by suggesting new design targets. We validate this method with a user study where participants build small prototypes that represent forest-replanting devices in a simulated fleet, with the goal of maximizing the fleet's effectiveness. The guided group followed the specifications given to them by the AI guide, while the unguided group chose their own specifications. While the mean effectiveness was similar between the groups, the guided group had a higher baseline effectiveness than the unguided group. Additionally, there was no significant reduction in the variety of the guided group's designs. This shows that AI-guided design can consistently help engineers find optimal solutions without negative effects on design exploration.
Continuing with the pioneering work by M. Graham Netting, the abundance of woodland box turtle (Terrapene carolina carolina) and wood turtle (Glyptemys insculpta) were opportunistically monitored for six decades (1959–2019) at the Powdermill Nature Reserve (PNR) in southwestern Pennsylvania. We found that both of the species grew slowly with delayed maturity and potentially long lives. Four wood turtles were recaptured 27–45 yrs after first marking, and five box turtles were 28–45 yrs post-marking. Both sexes of each species had high adult survivorship, with females exhibiting higher recapture probabilities. Although we found relatively few box turtles at PNR, their numbers were generally steady over six decades. Conversely, the wood turtle was frequently encountered in the 1960s but then had an apparent decline with no recovery. For example, of the 161 marked wood turtles, most (57.8%) were encountered during the first decade of study and only 15 were marked since 2004. We found no relationship between wood turtle numbers and major weather events such as severe cold periods or flooding. Thus, several factors may have coalesced to reduce their numbers at PNR: increased vehicle use of a road that bisects the property, habitat succession, stream pollution from prior coal mining, and collecting for the pet trade. As protective measures for these terrestrial turtle species, we suggest the construction of fences and under-road culverts for safe passage across roads, and the restoration of patches of open habitats across the property for turtle nesting.
La réalisation des opérations culturales en bonne condition est un enjeu majeur pour la réussite de l'intervention, et pour la préservation des sols face à l'augmentation du poids des agroéquipements.Un jour est disponible si les conditions sont favorables à la réussite de l'opération (travaillabilité) et le sol peut supporter le passage de la machine (traficabilité).Le projet J-DISTAS a pour ambition de construire un prototype de calcul des jours disponibles.Ce prototype interroge plusieurs modèles et bases de données pour calculer des indicateurs, qui sont ensuite agrégés selon des règles de décision construites par expertise.Le résultat obtenu tient compte de la travaillabilité et de la traficabilité : il permet de considérer le risque de tassement dans le dimensionnement du matériel ou l'évolution des systèmes de culture.La performance du prototype est en cours d'évaluation.
