Research Progress

• 07/17 -2020-
  Researchers propose new optimization method to improve global performances of collaborative robots

  This method takes the natural frequency, the Cartesian stiffness and the mass of the robot as optimization objectives, and can tackle the challenges in optimizing design of collaborative robots, such as the influence of time-varying kinematic configurations and optimization objectives, the calculated amount and the modeling precision of optimization model conflict with each other, and the influence of non-structural parameters on robot performances, etc.
• 07/07 -2020-
  New method helps keep an eye on electromagnetic coils degradation

  As the use of low-voltage (under 1kV) electrical rotating machines are more widely in new applications, especially in more-electric aircraft, the reliability of the low-voltage coil insulation systems in rotating machines has thus become a critical issue and requires technical-conditions-monitoring to avoid unexpected shutdown of machines that incorporate electromagnetic coils.
• 06/17 -2020-
  Wideband spectrum sensing technology boosts the next-generation intelligent radios

  Recently, the Industrial Communication and System-on-Chip (iComSoC) group at Shenyang Institute of Automation, Chinese Academy of Sciences, has studied the sensing function for the next-generation intelligent radio and proposed high performance wideband spectrum sensing algorithms with the exploitation of antenna cross-correlations and space-time information in multiple-input, multiple-output (MIMO) systems. The study was recently published in IEEE Transactions on Communications.
• 04/13 -2020-
  Researchers Propose Lifelong Machine Learning for Multi-view Task

  The Machine Intelligence research group at Shenyang Institute of automation (SIA), the Chinese Academy of Sciences (CAS) proposes a continual multi-view task learning model that integrates deep matrix factorization and sparse subspace learning in a unified framework, which is termed deep continual multi-view task learning (DCMvTL).
• 03/20 -2020-
  Porous material enables autonomous underwater vehicles to "swim" further

  Under the study, researchers conducted research on the mechanical characteristics and drag reduction mechanism of autonomous underwater vehicles using porous material, and performed preliminary comparative tests.
• 03/18 -2020-
  Progress made in smart grid data analysis research

  Under the leadership of professors WANG Zhongfeng and SONG Chunhe, the smart grid research group of Shenyang Institute of automation proposed an anomaly detection model suitable for unbalanced data distribution, aiming at the problem that the imbalance of abnormal data and normal data distribution results in significant reduction of anomaly detection accuracy in the process of smart grid data analysis.
• 03/18 -2020-
  New wheel-leg-crawler compound EOD robot developed by SIA

  In complex environments, Scorpio can use crawler walking to overcome obstacles and climb stairs; on flat city roads, it can use two-wheel self-balancing mode to walk, with high speed, high flexibility, low energy consumption and long battery life.
• 02/21 -2020-
  Infrared Stripe Correction Algorithm Proposed to Enhance Quality of Infrared Images

  Recently, researchers at Shenyang Institute of Automation of Chinese Academy of Sciences proposed an infrared image stripe non-uniformity correction algorithm that can use a single frame of infrared image to remove stripe non-uniformity and maintain edge details. Relevant research results are published in Applied Sciences-Base and Journal of the European Optical Society-Rapid Publications.
• 02/19 -2020-
  New Relay Node Placement Method Makes Transmission of WSNs More Timely and Reliable

  The research group on Industrial Wireless Networks at Shenyang Institute of Automation, Chinese Academy of Sciences (SIA) proposed a novel relay node placement method for wireless sensor networks (WSNs) which serves as a solid foundation to guarantee the real-time and reliable transmission of WSNs.
• 02/17 -2020-
  New method proposed to achieve better robot self-learning

  The research team proposed a new framework of incremental learning method based on Q-Learning and adaptive kernel linear (AKL) model. The framework allows robot to learn new behaviors without forgetting the previous ones. Under the new method, robot behaviors can be evaluated by means of autonomous learning and imitation learning, and the model structure and parameters can be changed in real time using a novel L2-norm kernel recursive least squares (L2-KRLS) algorithm.