"IROS 2019: Ninebot Robotics Emerges as Champion in Challenge Competition"

2019-11-05

The 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019) officially opened on November 4th. According to relevant statistics, the conference attracted over 4,000 professionals, top research team representatives, and industry professionals from around the world in the fields of robotics, automation systems, and artificial intelligence for exchange and communication. In addition to focusing on the latest research achievements in robotics, IROS also held robot competitions to showcase the charm of robots in a more direct and relaxed manner. IROS, jointly organized by IEEE, IEEE Robotics and Automation Society (RAS), IEEE Industrial Electronics Society (IES), Robotics Society of Japan (RSJ), Society of Instrument and Control Engineers (SICE), and New Technology Foundation (NTF), is the flagship international conference for robotics technology and intelligent systems. IEEE is a non-profit technical professional association with over 400,000 members in 160 countries/regions. It is a leading authority in various technological fields, from computer engineering, biomedical technology, and telecommunications to power, aerospace, and consumer electronics. It is one of the top international conferences in the field of intelligent robots and automation, and each IROS conference and exhibition has been highly successful, playing an important role in promoting technological development in related fields. IROS 2019 is the 32nd conference of IROS, where leaders from various industries, representatives of top research teams, and industry professionals will gather in Macau to explore the cutting-edge technology in the field of intelligent robots and systems, and share and discuss the latest developments in related fields. The conference will run until the 8th of this month and will include plenary sessions, keynote speeches, technical sessions, workshops and tutorials, forums, challenges, and exhibitions. The theme of this conference is "Robots Connecting People," aiming to promote the development of robot technology and make robots more human-like. The exciting events of the conference should not be missed. For more details, please click on the link at the end of the article to visit the conference website and explore the relevant content. The IROS robot challenge this year includes two competitions, and one of them is the Lifelong SLAM - Adaptive Scene Change Localization Algorithm Competition. SLAM, which stands for "Simultaneous Localization and Mapping," aims to enable robots to autonomously estimate their own position and orientation while moving, which is one of the core problems in the field of robotics. In simple terms, SLAM refers to the process of accurately establishing the correspondence between time and space when a robot enters a completely unfamiliar environment, and being able to answer questions such as: Where was I just now? Where am I now? What did I see? What is different from what I saw before? What should I do if I get lost? and so on. Usually, the study of SLAM focuses on the accuracy of robot positioning in static environments or in scenes with certain obvious dynamic characteristics, while ignoring the positioning failures and incorrect matching caused by scene changes. Therefore, this competition proposes the success rate of localization as an indicator, focusing on whether SLAM algorithms can stably recognize their own positions when there are changes in perspective, lighting, and scene layout, thus supporting long-term deployment of robots. At the challenge, the Segway Robotics team developed the Segway delivery robot S1, which performed exceptionally well in data collection. It is reported that the S1 is equipped with two sensors, the RealSense D435i camera and the RealSense T265 camera, to collect image data, both installed at a fixed height of about 1 meter. According to participating engineers, the Segway Robotics team integrated deep learning feature matching and scene relocation into the original positioning algorithm framework of the Segway delivery robot S1, greatly improving the robot's ability to continuously self-locate through vision. The comprehensive evaluation score ranked first, winning the championship in this challenge. It is understood that Segway Robotics is a subsidiary of Ninebot, dedicated to the business of robot innovation. The team has been deeply involved in the field of robotics since 2015, focusing on the research, production, and promotion of intelligent robots for individual consumers and the enterprise market. They have leading global capabilities in artificial intelligence algorithm development and hardware production. Segway Robotics has always attached great importance to the combination of academic frontiers and engineering implementation, and invests manpower and intern projects to participate in various academic conferences and competitions every year. In the recently concluded ICCV 2019 (International Conference on Computer Vision) held by Beihang University, SenseTime, Chinese Academy of Sciences, and Sun Yat-sen University, the intern team trained by Segway Robotics won second place in the project of "Interaction Detection between People and Objects in Real-Life Scenarios." This championship undoubtedly fully affirms the years of technical accumulation of Segway Robotics in the field of SLAM, proving the saying: "Where there's a will, there's a way." References: Wikipedia, IROS 2019 official website, Phoenix Finance, Xianji Network.

RECENT AI NEWS

RECENT AI TOOLS