ExpressMM: Expressive Mobile Manipulation Behaviors in Human-Robot Interactions

ExpressMM sits at the intersection of social robotics and mobile manipulation. While frontier labs (OpenAI, Figure, Tesla) are focused on the 'hard' problem of general-purpose manipulation and locomotion scaling, ExpressMM focuses on the 'soft' problem of legible communication via motion. With 0 stars and 4 forks, it is currently an early-stage academic reference implementation for a recently published paper. Its defensibility is low because it lacks a community, proprietary dataset, or hardware lock-in; it is primarily a set of algorithms that could be replicated by any HRI lab. The moat exists only in the specific domain expertise required to balance task efficiency with expressive legibility. Competitors include academic frameworks like Socially Aware Navigation (SAN) and various HRI libraries from CMU or Stanford. The main risk is that as Foundation Models for robotics (VLAs) mature, expressive behavior might emerge as a byproduct of training on human-centered data, potentially making manual expressive behavior frameworks obsolete within 1-2 years.