Publication Date:
2019-08-30
Description:
This work proposes a novel semantic perception system based on computer vision andmachine learning techniques. The main goal is to identify objects in the environment and extracttheir characteristics, allowing a dynamic interaction with the environment. The system is composedof a GPU processing source and a 3D vision sensor that provides RGB image and PointCloud data.The perception system is structured in three steps: Lexical Analysis, Syntax Analysis and finallyan Analysis of Anticipation. The Lexical Analysis detects the actual position of the objects (ortokens) in the environment, through the combination of RGB image and PointCloud, surveying theircharacteristics. All information extracted from the tokens will be used to retrieve relevant featuressuch as object velocity, acceleration and direction during the Syntax Analysis step. The anticipationstep predicts future behaviors for these dynamic objects, promoting an interaction with them interms of collisions, pull, and push actions. As a result, the proposed perception source can assignrelevant information to mobile robots, not only about distances as traditional sensors, but aboutother environment characteristics and object behaviors. This novel perception source introduces anew class of skills to mobile robots. Experimental results obtained with a real robot are presented,showing the proposed perception source efficacy and potential.
Electronic ISSN:
1424-8220
Topics:
Chemistry and Pharmacology
,
Electrical Engineering, Measurement and Control Technology
Permalink