The annual Robocon competition project allows HKU engineering students from different disciplines to design and fabricate innovative robots with an integration of various advanced technologies, including IoT sensors, AI, computer vision, and mobile computing. Besides, it provides hands-on experiences on product design, prototyping, CNC machining, design and fabrication of electric circuit and PCB, control program development, etc.
Month: September 2022
Autonomous excavators are an essential part of the goal of “building the robots that build the world”. One unique problem in autonomous excavation is how to deal with the granular materials like soils and sands, which is seldom studied in robotics. In this talk, Dr. Pan will present his team’s recent work about how to achieve efficient manipulation of soils by optimizing the trajectory of the excavator’s bucket, and how to enable the excavator to be aware of the objects buried in the soils by using a proximity sensing mechanism based on jamming in granular materials.
Laser scanning is used in advanced biological microscopy to deliver superior imaging contrast, resolution and sensitivity. However, it is challenging to scale up the scanning speed required for interrogating a large and heterogeneous population of biological specimens or capturing highly dynamic biological processes at high spatiotemporal resolution. Bypassing the speed limitation of traditional mechanical methods, free-space angular-chirp-enhanced delay (FACED) is an all-optical, passive and reconfigurable laser-scanning approach that has been successfully applied in different microscopy modalities at an ultrafast line-scan rate of 1-80 MHz. Optimal FACED imaging performance requires optimized experimental design and implementation to enable specific high-speed applications. In this protocol, we aim to disseminate information allowing FACED to be applied to a broader range of imaging modalities. We provide (i) a comprehensive guide and design specifications for the FACED hardware; (ii) step-by-step optical implementations of the FACED module including the key custom components; and (iii) the overall image acquisition and reconstruction pipeline. We illustrate two practical imaging configurations: multimodal FACED imaging flow cytometry (bright-field, fluorescence and second-harmonic generation) and kHz 2D two-photon fluorescence microscopy. Users with basic experience in optical microscope operation and software engineering should be able to complete the setup of the FACED imaging hardware and software in ~2-3 months.
Conventional mechatronic, hydraulic and pneumatic motors and actuators are used for large-scale robots from ≥10 cm to the human size. At the other, nanometric end of the length scale, nano-robots are powered by molecular motors. However, a number of applications in compact environments require robotic devices in the size range of 10 µm to 10 mm, but these are too small to be powered by the conventional mechatronic systems, and too large for molecular motors. Such a length scale ideally suits a few types of high-performance stimuli-responsive actuating materials that are emerging out of a very active research field in the past two decades, with examples including shape-memory polymers and metals, nanoporous noble metals, reactive polymers and liquid-crystal elastomers, carbon-based materials and transitional metal oxides. In addition to high actuating power densities, some of these materials also offer built-in sensory functions such as resistivity responses to mechanical, heat and humidity changes in the environment, and even energy generation capabilities. Integration of these materials and their signal flows in compact designs thus poses a novel strategy for robotics at the micro length scale. This talk will review some recent progress in this field.
In computer science, a graph is a network modeling objects and their unique interactions. The graph learning model is a specialized machine learning model that learns on graphs. Similar to traditional machine learning models, a well-performed graph learning model can capture the global data distribution with sufficient and unbiased training data. However, in a distributed subgraph system, most data owners only possess small amounts of the data (small subgraphs) in their local systems and can have unpredictable biases.
In this talk, the speaker will introduce this novel yet realistic setting – subgraph federated learning, which aims to let distributed data owners collaboratively train a powerful and generalized graph learning model without directly sharing their subgraphs. Towards this setting, two major techniques are proposed by the research team. (1) FedSage, which trains a GraphSage model based on FedAvg to integrate node features, link structures, and task labels on multiple local subgraphs; (2) FedSage+, which trains a missing neighbor generator along FedSage to deal with missing links across local subgraphs. Empirical results and theoretical analysis of proposed models respectively demonstrate the effectiveness and prove the generalization ability.
The vision of “Cyber-Physical Internet (CPI)” is to establish a new paradigm for sending and receiving manufactured goods just like sending and receiving instant messages over the internet using online chatting platforms. Four innovations are critical to achieve this ultimate vision: (1) digitization architecture for entangling the flows of information and materials into one flow of cyber-physical objects for manufacturing and logistics operations; (2) network services for configuring local aera network (LAN), wide area network (WAN) and catchment area network (CAN); (3) value mechanisms to motivate and facilitate participation and collaboration between multiple stakeholders including shippers, carriers, forwarders; and (4) decision analytics for synchronized logistics planning, scheduling and execution. These innovations are based upon some fundamental breakthroughs of CPI routers and TCP/PIP protocols that are yet to be developed.