Small Medium Enterprise (SME) scenario is quite different from common high-volume robotic applications. Many different types of technological operations are done in the same space and the working area of the robot should not be fenced in order to allow the operators to frequently access to it. Therefore, the need of a motion planning strategy to "real time" collision avoidance is mandatory. This work presents a new algorithm to avoid collisions between robots and human operators taking into account the complexity of an industrial scenario. The replanning and control problem has been split in two different modules: identification of safe evasive directions of motion suitable to avoid static obstacles (i.e., known in the off-line programming phase) and an enhanced Artificial Potential Field algorithm (APF) developed in order to impose the correct robot behavior to avoid dynamic obstacles (like human operator whose position is not predictable). In this work we assume to track the position of the human with suitable sensors.
Obstacle Avoidance Algorithm for Safe Human-Robot Cooperation in Small Medium Enterprise Scenario
N Pedrocchi;M Malosio;L Molinari Tosatti;
2009
Abstract
Small Medium Enterprise (SME) scenario is quite different from common high-volume robotic applications. Many different types of technological operations are done in the same space and the working area of the robot should not be fenced in order to allow the operators to frequently access to it. Therefore, the need of a motion planning strategy to "real time" collision avoidance is mandatory. This work presents a new algorithm to avoid collisions between robots and human operators taking into account the complexity of an industrial scenario. The replanning and control problem has been split in two different modules: identification of safe evasive directions of motion suitable to avoid static obstacles (i.e., known in the off-line programming phase) and an enhanced Artificial Potential Field algorithm (APF) developed in order to impose the correct robot behavior to avoid dynamic obstacles (like human operator whose position is not predictable). In this work we assume to track the position of the human with suitable sensors.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.