Algorithm helps robot fall gracefully

A new algorithm developed in Atlanta could reduce damage to costly robots.

ATLANTA, GEORGIA, USA (FEBRUARY, 2015) (SEHOON HA / KAREN LIU – GEORGIA TECH) – An algorithm has been developed at the Georgia Institute of Technology by PHD student Sehoon Ha and Professor Karen Liu which allows robots to fall over with minimal damage being caused.

According to Lui, the aim of the research is to enable robots to compute what happens when they fall so they can respond accordingly and minimise any potential impact.

Ha and Liu’s algorithm lets robots break a variety of fall types and allowed the robot’s use of its arms to protect its head, while sideways tumbles also helped avoid damage.

Their paper, ‘Multiple Contact Planning for Minimizing Damage of Humanoid Falls’, was presented in October at the International Conference on Intelligent Robots and Systems in Hamburg, Germany.