http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Xiaowei Xing,Dong Eui Chang 제어로봇시스템학회 2019 제어로봇시스템학회 국제학술대회 논문집 Vol.2019 No.10
Deep reinforcement learning trains neural networks using experiences sampled from the replay buffer, which is commonly updated at each time step. In this paper, we propose a method to update the replay buffer adaptively and selectively to train a robot arm to accomplish a suction task in simulation. The response time of the agent is thoroughly taken into account. The state transitions that remain stuck at the boundary of constraint are not stored. The policy trained with our method works better than the one with the common replay buffer update method. The result is demonstrated both by simulation and by experiment with a real robot arm.
MISFIRE FAULT DIAGNOSIS OF RANGE EXTENDER BASED ON HARMONIC ANALYSIS
Xiaowei Xu,Zhenxing Liu,Jiangdong Wu,Jiaming Xing,Xiaoqing Wang 한국자동차공학회 2019 International journal of automotive technology Vol.20 No.1
For an Extended-Range Electric Vehicle (E-REV), the misfire failure of the range extender can result in working condition deterioration, mileage decrease, vibration and noise increase, and other adverse consequences. The relationship of the shaft instantaneous angular speed (IAS) signal and the misfire fault of the range extender was studied by harmonic analysis. Based on the theory of shafting torsional vibration, the range extender dynamics model was developed to analyze the changing trend of the shaft IAS theoretically. Then, the shaft IAS signal under different working conditions was simulated using a developed range extender rigid-flexible hybrid multi-body dynamics model. It is found that the 0.5-order harmonic phase and the amplitude of range extender IAS can be used as the characteristic parameters in misfire fault diagnosis, which has been verified by experiments on an engine bench.
GRU-Attention based TD3 Network for Mobile Robot Navigation
Jiayao Jia,Xiaowei Xing,Dong Eui Chang 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11
In this paper, we propose a goal-oriented navigation reinforcement learning network called GRU-Attention based TD3 network, which takes lidar measurements, the distance between agent and target, and yaw toward the target as state inputs. The policy in the network outputs continuous actions consisting of forward velocity and yaw angular velocity. Our proposed network can perform obstacle-avoidance navigation without prior knowledge of the environment. We train our network in a simulation environment. To show that our proposed network is better in navigation tasks, we compare its performance with two other networks: the pure TD3 network and the GRU-based TD3 network in several different simulation worlds. The experiments show that the mobile robot with our proposed network can bypass the obstacles safely and arrive at the goal positions as fast as possible. The supplementary video is given at: https://youtu.be/HkqUZSsT5a0. The implementation is made open source at: https://github.com/Barry2333/DRL_Navigation.git.
Autonomous Drone Surveillance in a Known Environment Using Reinforcement Learning
Mengyi Gao,Xiaowei Xing,Dong Eui Chang 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11
We utilize deep reinforcement learning to develop both single-agent and multi-agent methods that can accomplish autonomous drone surveillance tasks in a known indoor environment in this research. We combine the benefits of both visual and obstacle information to boost efficacy while ensuring low time consumption. And we devise a separate reinforcement learning training and test technique that both enhance training efficiency and ensure task completion. This method also creates a new field for sim-to-real transfer. Our experimental results show that the trained agents can detect all targets at a relatively fast speed while maintaining a high level of security, and the patrol completion rate is more than 98% in both single-agent and multi-agent tasks.
Sanrong Li,Jing Ma,Caiying Hu,Xing Zhang,Deyong Xiao,Lili Hao,Wenjun Xiao,Jichun Yang,Ling Hu,Xiaowei Liu,Minghui Dong,Duan Ma,Rensheng Liu 한국유방암학회 2018 Journal of breast cancer Vol.21 No.3
In this study, we used next-generation sequencing methods to screen 300 individuals for BRCA1 and BRCA2. A novel mutation (c.849dupT) in BRCA2 was identified in a female patient and her unaffected brothers. This mutation leads to the truncation of BRCA2 functional domains. Moreover, BRCA2 mRNA expression levels in mutation carriers are significantly reduced compared to noncarriers. Immunofluorescence and western blot assays showed that this mutation resulted in reduced BRCA2 protein expression. Thus, we identified a novel mutation that damaged the function and expression of BRCA2 in a family with breast cancer history. The pedigree analysis suggested that this mutation is strongly associated with familial breast cancer. Genetic counsellors suggest that mutation carriers in this family undergo routine screening for breast cancer, as well as other malignancies, such as prostate and ovarian cancer. The effects of this BRCA2 mutation on drug resistance should be taken into consideration during treatment.