• The Journal of Korea Robotics Society
• /
• v.18 no.2
• /
• pp.143-154
• /
• 2023

In this study, a fast motion planning method for the swing motion of a 6x6 wheel-legged robot to traverse large obstacles and gaps is proposed. The motion planning method presented in the previous paper, which was based on trajectory optimization, took up to tens of seconds and was limited to two-dimensional, structured vertical obstacles and trenches. A deep neural network based on one-dimensional Convolutional Neural Network (CNN) is introduced to generate keyframes, which are then used to represent smooth reference commands for the six leg angles along the robot's path. The network is initially trained using the behavioral cloning method with a dataset gathered from previous simulation results of the trajectory optimization. Its performance is then improved through reinforcement learning, using a one-step REINFORCE algorithm. The trained model has increased the speed of motion planning by up to 820 times and improved the success rates of obstacle crossing under harsh conditions, such as low friction and high roughness.

• Environmental Analysis Health and Toxicology
• /
• v.15 no.4
• /
• pp.131-137
• /
• 2000

The release of hazardous waste materials into the environment poses serious risks in humans and ecosystems. The risk assessment of environmental pollutants including hazardous chemicals requires a comprehensive measurement of hazard and exposure of the chemicals that can be achieved by toxicity evaluation using a biological system such as biomarkers. In this report we have tried to develop a biomarker used to elucidate a molecular basis of, and to monitor abnormal behaviors caused by diazinon in Japanese medaka (Oryzias latipes) as a model organism. First, an attempt was made to clone tyrosine hydroxylase gene from Japanese medaka that would be a candidate for a biomarker for neuronal modulations and behaviors. For monitoring experiments at behavioral and molecular biological levels, the fish were treated under different sublethal conditions of diazinon and their behavioral responses were observed . In this study we have successfully cloned a partial TH gene from the medaka fish through PCR screening of an ovary cDNA library. DNA sequencing analysis revealed that the amplified fragment was 327 bp encoding 109 amino acids. Comparing the DNA sequence of medaka TH with other species, TH gene revealed the DNA sequence was completely identical to that of rat TH. In the RT-PCR, 330 Up of mRNA was consistently amplified in all the treated samples including control There were no significant differences in the TH expression level regardless of treating concentrations (1∼5,000 ppb) and time (0∼48 hr) The reason appeared to be that RT-PCR was not performed using through a quantitative analysis normalized against an actin gene expression. Organ or tissue - specific detection of TH activity and mRNA as biomarkers will be a useful monitoring tool for neurobehavioral changes in fish influenced by toxic chemicals. Furthermore, quantitative analysis of locomotive patterns and its correlation with the neurochemical and molecular data would be highly useful in measuring toxicity and hazard ofvarious environmental pollutants.