As a modular and reactive control approach, constraint-based programming helps us to formulate and solve complex robotic tasks in a systematic way. In different fields ranging from industrial manipulators to humanoids, robots are supposed to work in an uncertain environment. However, how to address uncertainties is missing in the state-of-the-art of different constraint-based programming frameworks. In this paper, we introduce a Second Order Cone Programming (SOCP) approach to integrate constraints with norm bounded uncertainties. The proposed SOCP is convex and through simulations with controlled uncertainty level, we can clearly tell that the proposed approach guarantees the constraints satisfaction compared to the state-of-the-art.
QC 20180522