IEEE CIS ISATC Robotics Task Force

Leadership

Chair

Fei Zhou
China Electric Power Research Institute, China
zhoufei@epri.sgcc.com.cn

Vice-Chairs

Kaoru Hirota
Tokyo Institute of Technology, Japan
hirota@jsps.org.cn

Wei Chen
Westlake University, China
chenwei06@westlake.edu.cn

Xi Chen
GEIRI North America, USA
xc@ieee.org

Members

  • Aysegul Ucar, Firat University, Turkey
  • Chee Seng Chan, University of Malaya, Malaysia
  • Chu Kiong Loo, University of Malaya, Malaysia
  • Dongbing Gu, University of Essex, UK
  • Feng Shan, Southeast University, China
  • Georgy Sofronov, Macquarie University, Australia
  • Guan Wang, Learnable.ai Co., China
  • Honghai Liu, University of Portsmouth, UK
  • Huiyu Zhou, Queen’s University Belfast, UK
  • Jason Gu, Dalhouise University, Canada
  • Jian Dong, Information Technology Research Center of China Electronics Standardization Institute
  • Jian Li, State Grid Jibei Electric Power Co., Ltd, China
  • Jianxin Li, Beihang University, China
  • Kheng Lee Koay, University of Hertfordshire, UK
  • Kyung-Joong Kim, Sejong University, Korea
  • M. Reza Emami, University of Tronto, Canada
  • Min Jiang, Xiamen University, China
  • Naoki Masuyama, Osaka Prefecture University, Japan
  • Naoyuki Kubota, Tokyo Metropolitan University, Japan
  • Paolo Remagnino, Kingston University London, UK
  • Qing Wang, State Grid Corporation of China Co., Ltd, China
  • Richard J. Duro, Universidade da Coruña, Spain
  • Rongwei Ren, Central Standardization Department of the Beijing Humanoid Robot Innovation Center, China
  • Rui Xin, State Grid Hebei Electric Power Co., Ltd, China
  • Ryad Chellali, Nanjing University of Technology, China
  • Samir Garbaya, Arts et Metiers ParisTech, France
  • Simon X. Yang, University of Guelph, Canada
  • Sung-Bae Cho, Yonsei University, Korea
  • Tetsuya Ogata, Waseda University, Japan
  • Tian Lan, Global Energy Interconnection Research InstituteEurope GmbH, Germany
  • Wanyuan Wang, Southeast University, China
  • Wei Jiang, State Grid Digital Technology Holding Co.,Ltd, China
  • Weilai Jiang, Beijing Humanoid Robot Innovation Center, China
  • Yong Deng, State Grid Fujian Electric Power Co., Ltd, China
  • Zhaojie Ju, University of Portsmouth, UK
  • Zhen Li, Beijing Institute of Technology, China

Mission

Scope

  1. Innovation from a Theoretical Perspective
    1. Brain-inspired Adaptation Mechanism Build a dynamic learning framework based on spiking neural networks to achieve biological temporal encoding of equipment state perception and fault patterns, breaking through the limitations of traditional static models in characterizing the gradual deterioration of equipment.
    2. Multimodal Cognitive Architecture Develop a brain-inspired heterogeneous cognitive system that integrates visual, thermal, and electromagnetic sensing. Establish a dynamic knowledge graph containing various equipment characteristic parameters to support multi - dimensional state interpretation of key equipment such as transformers and GIS.
    3. Dynamic Knowledge Evolution Introduce an online learning mechanism driven by neural plasticity, enabling the robot system to continuously update the equipment deterioration model in complex scenarios such as UHV converter stations and adapt to the rapid development needs of the new - type power system.
  2. Methodological Innovation
    1. Neuromorphic Computing Engine Develop a dedicated brain - inspired chip that supports event - driven computing to achieve the low - power goal of substation/converter station inspection robots, meeting the 7×24 - hour continuous monitoring needs of outdoor equipment.
    2. Heterogeneous Computing Integration Construct a cognitive decision - making system with a hybrid ANN - SNN architecture. Compress the parameter scale of large models through pulse temporal encoding technology to achieve real - time response after the generation of control instructions.
    3. Biomechanical Coupling Develop a brain - inspired controller that mimics the motor control mechanism of the human cerebellum to improve the trajectory tracking accuracy of live - working robotic arms under strong electromagnetic interference, breaking through the stability bottleneck of traditional control in unstructured environments.
  3. Technological System Innovation
    1. Cross - modal Cognitive Interface Develop a brain - inspired sensing array that supports the coupled analysis of multiple physical fields to achieve synchronous detection and correlated reasoning of power signals such as SF6 gas leakage and partial discharge ultrasound.
    2. Autonomous Evolution System Construct a two - way optimization framework that mimics the brain's dopamine reward mechanism, enabling inspection robots to autonomously optimize path - planning algorithms when performing tasks such as temperature measurement of GIS equipment (with a task efficiency improvement of over 15%).
  4. Key Research Directions
    1. Cross - domain Mapping between Brain - inspired and Equipment Systems Solve the difficult problem of equivalent conversion between neural pulse encoding and equipment status parameters. Establish a library of pulse spatiotemporal patterns that includes multiple typical fault modes.
    2. Mechanism for Balancing Energy Efficiency and Accuracy Break through the mutually exclusive contradiction between brain - inspired sparse computing and detection accuracy. Control the power consumption of computing units within a few watts while maintaining a defect recognition rate of over 99%.
    3. Human - machine Hybrid Intelligence Enhancement Design a standard - compliant cognitive collaboration interface to achieve millisecond - level conversion of dispatchers' natural language instructions into brain - inspired control signals. Construct a closed - loop optimization system for human experts and machine intelligence.

2025 Planned Activities