Reproduce “Adaptive Shared Control with Online Bounded-Rational Human Behavior Estimation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Henry Ascencio Trejo, Roel Pieters, Gokhan Alcan
The controller estimates human behavior by comparing state motion with a finite bank of policies, then chooses assistance by minimizing expected cost across the estimated distribution. Under a quadratic terminal-value approximation, Euler propagation, and strict convexity, it has a closed-form action. In a 30-second simulation, robot cost was 112.42, below two baselines; evaluations were simulations with fixed tasks and intentions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.