The sustainable cultivation of organic vegetables and the associated problem of weed control has been a current research topic for some time. Despite this, the use of chemical and synthetic pesticides increases every year. This is to be solved with the help of an automated robot system. The current version of the weeding robot uses GPUs to execute the inference phase. This requires a lot of energy for an 8-track robot. To enable autonomous solar operation, the system must be made more energy efficient. This work aims to evaluate possible approaches and the current state of research on implementing convolution neural networks on low power embedded systems. In the course of the work, the technical feasibility for the implementation of CNNs in FPGAs was examined, in particular, following the example of a feasibility analysis. This paper shows that the acceleration of convolution neural networks using FPGAs is technically feasible for use as detection hardware in the weeding robot. With the help of the current state of research and the existing literature, the optimization possibilities of the hardware and software have been evaluated. The trials of different networks on different hardware accelerators with diverse approaches were investigated and compared.

Summary: Reasons for performing study: Exercise depletes muscle glycogen stores, which could subsequently impair performance. Muscle glycogen replenishment is determined by substrate availability. Objectives: To study the effects of feeding meals of varying glycaemic responses on blood concentrations of substrates used for glycogenesis in horses with exercise‐induced glycogen depletion. Methods: In a 3‐way crossover study, 7 horses received each of 3 isocaloric diets for 72 h after undergoing glycogen‐depleting exercise: 1) a high soluble‐carbohydrate diet, which induced a high‐glycaemic (HGI) response; 2 and 3) a low soluble‐carbohydrate or a mixed soluble‐carbohydrate diet (control group), which both induced a similar low‐to‐moderate glycaemic (LGI) response. Muscle biopsies and venous samples were collected before and up to 72 h after exercise. Results: Feeding HGI diet resulted in a higher (P<0.001) rate of muscle glycogenesis over 72 h compared to LGI diets. Plasma glycerol, triglyceride, lactate, serum NEFA and total protein concentrations, and haematocrit were significantly (P<0.001) higher after compared to before exercise. Whereas no significant overall dietary effect was observed on these metabolites over 72 h, there was a tendency for glycerol, NEFA and triglyceride concentrations to be lower for LGI compared to HGI diets over 6 h after exercise (P<0.05; 1, 6 and 4–6 h after exercise, respectively). Conclusions: These data suggest that horses fed LGI meals after exercise had limited lipid utilisation without any significant shift of substrate utilisation toward gluconeogenesis, which could have contributed to the slower rate of muscle glycogenesis compared to horses fed HGI diets.