1.5: Challenges of Autonomous Manipulation

Last updated
Save as PDF

Page ID: 14771

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

Think about the last time you worked with your hands. This includes typing on your keyboard, writing on a piece of paper, sewing a button onto a shirt, and using a hammer or a screwdriver. You will notice that these activities require a wide range of dexterity, that is the ability to manipulate objects with precision, a wide range of forces, and a wide range of sensorial capabilities. You will also notice that some tasks go beyond your capabilities, such as putting yarn through a hole in fabric, grasping a screw, or driving a nail into a piece of wood, but can be easily solved with the right tool.

So far, robotic hands are far from reaching the dexterity of a human hand. Yet, with the right tool (called “end-effector” in robotics speech) some tasks can be solved even better, that is faster and more precisely, than by humans. As for solving a mobile robotics problem, manipulation problems require you to think about the right mix of reasoning and mechanism design. For example, grasping tiny parts might be impossible with tweezers, but really easy when using a sucking mechanism. Or, picking up a test tube that is hardly visible with the robots’ sensors can be picked up almost blindly when using a funnellike mechanism at your end-effector. Unfortunately, these tricks will most likely limit the versatility of your robot, requiring you to think about the problem and the users’s need as a whole.