Human inspired lighting control in robot systems with the intention of glare avoidance

The performance of robot vision perception, and thus the quality of information obtained from the environment, can be greatly affected by the conditions of surrounding lighting. This paper presents some human-inspired strategies for lighting control in a robot system for best scene interpretation, where the main intention is to avoid possible glares or highlights occurring in images. It firstly compares the characteristics of human eyes and robot eyes. Then some evaluation criteria are addressed to assess the lighting conditions. A bio-inspired method is adopted to avoid the visual glare which is caused by either direct illumination from large light sources for indirect illumination reflected by smooth surfaces. Appropriate methods are proposed to optimize the pose and optical parameters of the light source and the vision camera.