Grant: Discovery Grants Program - Individual

We are shifting from an era of mass production to an era of mass customization, and robots will play a key role in that shift. However, when it comes to the robots themselves, don't expect customized robots for every application anytime soon. Indeed, an application might be to displace a tool along three axes and orient it about two axes only, but the user will still end up acquiring a six-axis robot arm for various practical reasons. Or a user might prefer a specific industrial robot for some reason, but because it lacks a few centimeters of reach, the user will end up installing that robot on a linear guide. Thus, in many cases we end up with so-called kinematically redundant robot systems (KRRS) that have more degrees of freedom than required by the task to be accomplished. Sometimes, it may even be wise to go for a KRRS, such as a 7-axis robot arm, right from the start just in case we end up needing more flexibility. While 7-axis robot arms, like the Canadarm, have existed for decades, these robots became widely popular only recently, mostly with the advent of so-called collaborative robots (cobots). Indeed, 7-axis cobots are now commercially available from ABB, Rethink Robotics, Kinova, Franka Emika, Productive Robotics, and Kassow Robots, to name a few. The problem with 7-axis robot arms, and KRRS in general, is that these systems are even more complex to use than their non-redundant counterparts. And almost always, the user is provided no assistance in selecting the redundant motion parameters, such as the so-called arm angle in 7-axis robots. Therefore, further research efforts are needed in order to render the use of KRRS more intuitive and more beneficial. The general long-term objective of the proposed research program is the kinematic study and improvement of existing-and the design of better-KRRS. One of the most challenging research axes will be the study of the kinematics of KRRS, and specifically the analyses of the singularities and inverse kinematics of 7-axis robot arms. Better insights into these two complex problems will eventually allow users to be able to get more from these arms. In the case of two-module KRRS, methods for computing the relative workspace of one end-effector with respect to the other will be developed. A second research axis will be the development of techniques for the optimal management of the redundant motion parameters, for various application scenarios. One of these scenarios is collaborative robotic applications, where the redundancy can be used to reduce the risks for the human operator. A third research axis will be the development of self-calibration methods for KRRS, based on an in-depth knowledge of the redundant self-motion. These methods will improve the accuracy of KRRS using simple linear gages and kinematic couplings, instead of expensive laser trackers. Finally, a fourth axis of the proposed research program will be design of novel KRRS with simpler kinematics and optimal design.