E-mail： artheru@sjtu.edu.cn
Mobile： +86-15682335702
Birthday： September 7th, 1990
Repo：https://git.lessokaji.com
Introduction
Liu Shicong, Ph.D. from Shanghai Jiao Tong University, is an expert in the R&D ofindustrial mobile robots. He has innovatively developed the MDCS full-stack robotsoftware system (including OS, perception, control and fleet management), CycleGUIturn-based 3D user interface framework, and more. Currently, his primary focus is onresearch and development of artificial intelligence technology applications in theindustrial field, such as visual navigation, robot control, robot cluster scheduling,programming languages, etc. His expertise drives numerous robotics industrycompanies in China.
During his Ph.D. studies, his main research direction was foundational AI algorithms,including computer vision, deep learning, big data processing, and automaticprogramming. To date, he has published a number of academic papers in topinternational journals and conferences such as T-MM, ICDM, ICME, IJCNN, and hasobtained several patents. He was recommended to Shanghai Jiao Tong University in2009 through the NOIP Informatics Competition and obtained a bachelor's degree in013. He pursued a Ph.D. immediately after, and obtained the degree in June 2019.
His current R&D efforts primarily revolve around SLAM (Simultaneous Localization2and Mapping) technology with vision and laser, vision-based semantic space perceptionand navigation, intelligent agent decision-making based on AIGC technology, industrialAGI, and industrial AR/VR engines.
Education
2009-2013： Shanghai Jiao Tong University, School of Electronic Information andElectrical Engineering, Computer Department. Obtained a Bachelor's degree.
2013-2019： Shanghai Jiao Tong University, School of Electronic Information andElectrical Engineering, Computer Department. Obtained a Ph.D. degree. Focus:Applications of deep learning in the vision field and content-based massiveinformation acquisition. Thesis title: "Structured Encoding Learning and Indexing ofLarge-Scale High-Dimensional Information."
Experience
From September 2016 to Present
Founder of Lessokaji Co., Ltd. Developed SLAM algorithm based on ground texture or laser or panoramic vision for multiple clients. Developed the completeAGV software system (MDCS system), the 3D warehouse control system, and theindustrial 3D and UI engine (CycleGUI).
January 2017 to September 2020
CTO at Shanghai Shuchao Intelligent Technology Co., Ltd. Accepted investmentfrom Jinshan Petrochemical Logistics Co., Ltd. Developed Robot ProcessAutomation (RPA) systems and software tools, provided information systemintegration services to clients, and developed software systems like WMS, OA,Quality Management System, etc.
November 2020 to June 2021
CTO at Huaxiao Robotics/CSG Group. Introduced and implemented the full suiteof MDCS AGV technology.
From June 2021 to Present
CTO/Cofounder at Fairyland Technology. Implemented the full suite of MDCSAGV technology and spearheaded engineering projects.
Papers and patents:
Generalized Residual Vector Quantization and Aggregating Tree for Large ScaleSearch IEEE Transactions on Multimedia, 2017
Quantizable Deep Representation Learning with Gradient Snapping Layer for LargeScale Search. International Conference on Multimedia and Expo, 2017
Space Shuttle Model: A physics inspired method for learning quantizable deeprepresentations. International Conference on Multimedia and Expo, 2017
Generalized residual vector quantization for large scale data. IEEE InternationalConference on Multimedia and Expo (ICME), 2016
Aggregating Tree for Searching in Billion Scale High Dimensional Data. IEEE 16thInternational Conference on Data Mining Workshops (ICDM), 2016
Accelerated distance computation with encoding tree/forest for high dimensionaldata. International Conference on Machine Learning and Cybernetics (ICMLC), 2016CN201710264333.9 - A visual navigation method based on ground image texture.
CN201710244401.5 - A method and device for implementing applications.
CN201810174438.X - A method for function program persistence, electronic device,and storage medium.
CN201810539970.7 - A method for laying and optimizing AGV visual tracks.
CN201810539969.4 - A method for acquiring real-time operating posture basedon an AGV cart.
CN111415390A - A location navigation method and device based on groundtexture.
CN111191759A - A method for generating fiducial marker and a GPU-basedlocation and decoding method.
CN202010918896.7 - A positioning method based on point cloud transformationmatching.
CN202011162835.9 - A positioning method for mobile devices based on multi-sensor coupling.
CN202111202799.9 - A route planning and management system and method formobile machinery equipment.
CN202310547441.2 - A rolling-wing device and aircraft.Additionally - Holds multiple software copyrights.
