eduzhai > Physical Sciences > Physics Sciences >

Deep Learning for Robotics

  • Save

... pages left unread,continue reading

Document pages: 14 pages

Abstract: The application of deep learning to robotics overthe past decade has led to a wave of research into deep artificial neuralnetworks and to a very specific problems and questions that are not usuallyaddressed by the computer vision and machine learning communities. Robots havealways faced many unique challenges as the robotic platforms move from the labto the real world. Minutely, the sheer amount of diversity we encounter inreal-world environments is a huge challenge to deal with today’s roboticcontrol algorithms and this necessitates the use of machine learning algorithmsthat are able to learn the controls of a given data. However, deep learningalgorithms are general non-linear models capable of learning features directlyfrom data making them an excellent choice for such robotic applications.Indeed, robotics and artificial intelligence (AI) are increasing and amplifyinghuman potential, enhancing productivity and moving from simple thinking towardshuman-like cognitive abilities. In this paper, lots of learning, thinking and incarnation challengesof deep learning robots were discussed. The problem addressed was roboticgrasping and tracking motion planning for robots which was the most fundamentaland formidable challenge of designing autonomous robots. This paper hope to provide the reader an overview of DL and robotic grasping, also the problem oftracking and motion planning. The system is tested on simulated data and realexperiments with success.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×