The present study, black soldier flies (Hermetia illucens) fermented using lactic acid bacteria were powdered without defatting and added to 3% or 5% to make pig feed. Weaning piglets were fed 3% (T3) or 5% (T5) feed powdered with Hermetia illucens for 5 months and the efficacy of the feed was investigated. The results of measuring body weight gain over 5 months after adding 3% (T3) or 5% (T5) of Hermetia illucens powder to the feed of weaned piglets showed significant weight gain in the T5 group compared to the control group. The added feed to Hermetia illucens powder did not show toxicity, and analysis of its effect on blood properties showed that white blood cell levels tended to increase in the T3 or T5 group compared to the control group.The only increase in white blood cell count was a change within the normal range. As a result of analyzing the effect of the level of addition of Hermetia illucens powder on feces, the effect of liquid reduction showed excellent results in the T3 treatment group and maintained the best form of feces. In this study, the thawing loss in the control group was 6.66%, and the T3 group with added powder to Hermetia illucens showed a significant decrease of 5.03%, and the T5 group also showed a decrease of 5.61%. Therefore, it was demonstrated that additive feed for Hermetia illucens reduced thawing loss, affected the water holding capacity of meat, and played an important role in maintaining the taste of meat. Moreover, the results of carcass grade showed a tendency for one grade to increase in the T3 and T5 groups fed additive feed to Hermetia illucens compared to the control group. In conclusion, the results of this study suggest that feed supplemented with Hermetia illucens is effective in influencing the weight gain of pigs, reducing the liquid content of feces, and increasing carcass grade.

In this paper, we present a learning platform for robotic grasping in real world, in which actor-critic deep reinforcement learning is employed to directly learn the grasping skill from raw image pixels and rarely observed rewards. This is a challenging task because existing algorithms based on deep reinforcement learning require an extensive number of training data or massive computational cost so that they cannot be affordable in real world settings. To address this problems, the proposed learning platform basically consists of two training phases; a learning phase in simulator and subsequent learning in real world. Here, main processing blocks in the platform are extraction of latent vector based on state representation learning and disentanglement of a raw image, generation of adapted synthetic image using generative adversarial networks, and object detection and arm segmentation for the disentanglement. We demonstrate the effectiveness of this approach in a real environment.