An AOP-Based Robot Behaviors Safety Checking Method

Abstract

A robot is a typical cyber-physical system whose behaviors are driven and controlled by software which are highly related with robot safety. In order to ensure the safety of robot behaviors, programmers need to embed safety logic into the programs in charge of behaviors. This paper proposes an AOP-based robot behaviors safety checking method to ensure robot behaviors safety at the programming level, which treats the safety constraints as a crosscutting concern to be automatically and transparently managed, thus enabling programmers to focus on the design of the robot behaviors themselves rather than safety. This paper considers two types of safety issues, stability and collision, and proposes corresponding algorithms for ensuring them at runtime. Experiments with real humanoid robots show that the method can effectively ensure the safety of robot behaviors with respect to the specified physical constraints.