UK RAS STEPS members participated in a three-day Micromaze Robot Hackathon at the University of York. Teams were challenged to design, build, and program autonomous robots to navigate complex mazes. Participants utilized a Raspberry Pi Pico W platform, MicroPython, and various sensors for collision avoidance. The challenge evaluated robot design, code quality, and wireless control capabilities. The event also aimed to foster future robotics education and outreach opportunities.
Humanoid robots are studied in tasks like traversing sparse 3D structures by executing agile whole-body motions. Historical use of robots was documented in urban search and rescue operations following the September 11th attacks. A foundation model like UnifoLM-WLA-1.0 achieved state-of-the-art results by coordinating desktop and whole-body manipulation. Compliance is crucial in physical interaction, demonstrated by work on hybrid impedance-admittance control for aerial robots. Achieving agile and generalized legged locomotion requires integrating perception and control, as detailed in a novel reinforcement learning framework.