 Hello everyone, I'm Wang Shangguang from Beijing University of Posts and Telecommunications, or BUPT. Zhang Xi from Huawei Cloud and I will share with you a report regarding the application of the cloud-native edge computing platform KubaEdge in aerospace. This report includes four parts. I'd like to introduce two of them, the status and challenges of satellite computing and the Sky Computing Constellation Program. Zhang Xi will present you the last two parts. As we know, satellites are becoming increasingly universal, intelligent and software-based. First, satellites increasingly tend to use universal technologies and components. For example, some industrial-grade components are used in satellites, especially in positioning satellites. Another trend is intelligence. GPU and NPU computing is now supported by satellite payloads. In addition, satellites become increasingly software-based. Satellite payloads, including systems and software, can be called through APIs. The onboard computer enables interconnection between software using network ports. Another feature of satellites is that networking is becoming more convenient, diverse and converged. One of the challenges is that the amount of data generated by satellites in orbit has exceeded the amount of data generated by satellite communications since 2020. In the past, satellites were mainly used to transmit data and network signals for communications. However, since 2000, both the know of satellites in orbit and the data generated have skyrocketed. The amount of data generated by satellites has actually exceeded that transmitted by satellites. Therefore, real-time in-orbit processing is required to clear redundant and useless data, shorten the response time and reduce the pressure or workload of network link transmission. In addition, services such as remote sensing, emergency response and disaster forecasting are eager for in-orbit computing and processing to improve the response speed and prediction accuracy. To meet these requirements, the Shenzhen Institute of BUPT established the Interplanetary Network and Intelligent Computing Lab in June 2020. The lab conducts satellite-related research, including interplanetary networks, satellite networks and distributed AI computing. For example, in March 2021, we completed the encapsulation of the Satellite Ground Collaboration Platform. In July, we encapsulated the in-orbit, intelligent computing and service platform. These two platforms, along with Kuba Edge, were integrated into the satellite launched at Jiuquan Satellite Launch Center on December 7. In August, we deployed a lightweight 5G core network on a satellite for the first time, realizing interconnection and communication with terrestrial 5G private networks at the signalling layer, as well as traffic splitting for video calls and edge computing. In September, we started to develop our first satellite, BU-PT1. In October, we started the Sky Computing Constellation Program with SpaceTie Co-Limited. The purpose of the Sky Computing Constellation Program is to build a constellation as the foundation of intelligent and comprehensive digital infrastructure, providing support to technologies such as 6G and satellite internet, and implementing an open-source in-orbit platform of aerospace computing for the scientific researchers around the world. Currently, 80% of the land and 90% of the sea on the earth are not covered by network signals. For example, underdeveloped areas such as Africa cannot benefit from ICT development. We hope that our constellation will provide a platform for more researchers or users to carry out open-source experiments benefiting more people. The program includes six key tasks. The first one is to conduct a systematic experiment of the 6G core network that uses a distributed architecture based on cognitive services. The second task is to carry out an experiment on the Internet of Data, or IOD, for satellites. The third one is to develop a next-generation satellite OS for interconnection between the satellite platform and the payload system. We also want to carry out tests on certain devices and effective payloads in space. In addition, we'll conduct experiments to verify distributed capabilities such as machine learning, federated learning, and onboard service capability openness. A satellite carrying our computing platform was successfully launched on December 7, 2021, for a preliminary verification experiment. In May 2022, we're going to launch the first satellite of the Sky Computing Constellation Program, BU-PT1. By 2023, all satellites of the program will have been launched and networked. Now let me take over the presentation to introduce how CUBE Edge works in this program. CUBE Edge is a project designed for edge computing and edge collaboration. It was open-sourced in 2018. From 2018 to 2019, we focused on developing CUBE Edge's capabilities. In 2020, the focus was to collaborate with peripheral ecosystems. In 2021, we have made some in-depth attempts at a wide range of fields, including AI, robotics, wireless, and MEC. CUBE Edge now is an incubating project of CNCF with more than 50 organizations and 800 contributors, including 220 code contributors around the world. CUBE Edge plays an important role in the cloud-native satellites. First, CEDNA, a sub-project of edge collaboration AI in CUBE Edge, is used to build multi-model collaborative inference on the ground and satellite, and incremental model training on the cloud that is on the ground. In this way, a small model is used on the satellite and a large model is used on the ground, so that the satellite requires few resources to achieve better AI inference effects. In addition, the device mapper of CUBE Edge is used to model and manage sensors of the satellite in a unified manner, allowing management personnel on the ground to obtain the working status of onboard devices in real-time. All of these communicate with each other through a highly reliable Cloud Edge channel established by CUBE Edge. A Kubernetes application model is also used to realize unified lifecycle management of applications on the satellite. Next, we are going to carry out experiments on satellites with two plans. The first plan is to make satellite in orbit computing more intelligent in two ways. One way is to further accelerate the inference at the edge by optimizing the inference algorithms, so that we can reduce the amount of data transmitted from the satellite to the ground. In addition, we will introduce the lifelong learning method to better process the heterogeneous data at the edge and better establish and share the knowledge base across satellites, achieving intelligent collaboration among satellites. The other plan is to build a cloud-native aerospace computing platform. We will define the roles of satellites based on their devices and locations. For example, there are satellites responsible for communication with the ground, for data and image collection, or for intelligent computing. These satellites are dynamically networked to negotiate and collaborate with each other. This process needs more intelligent networking and better service collaboration, which are to be explored and realized by CUBE Edge. Finally, thank you and you are welcome to join the community and participate in experiments and discussions on satellite in orbit computing and aerospace computing.