Hyper-Actuated Flexure-Link Servo Mechanisms

Project: Research council

Project Details


Conventional mechanisms used in robotics and automated machinery have joints with bearing parts that collide, roll and slide against each other. The associated interaction forces have an effect on small-scale motion that limits achievable accuracy when motion is controlled automatically using motors or other forms of actuation. This impacts negatively on the quality and efficiency of various industrial processes, including the assembly and inspection of manufactured products. If bearings are replaced by compact deformable structures acting as pseudo-joints then small-scale motion behaviour becomes more predictable and precise. Additional advantages are derived from the use of parts that do not rub or wear or require lubrication. A new issue that arises, however, is that flexible joints introduce additional ways in which a mechanism can move and vibrate. These motions must be regulated through suitable actuation and control schemes. Research is needed on how best to design pseudo-jointed structures to achieve accurate control of motion, not only for precise positioning but also during rapid large-scale configuration changes, without causing unwanted oscillations or instabilities. How to apply actuation forces to a mechanism structure is an important consideration here, but so also is the creation and use of mathematical models to: 1. Design mechanisms for an optimized balance of speed, precision and range of motion. 2. Develop algorithms that will run on a computer to regulate actuation forces and thereby achieve precise control of motion. This research will find new ways to solve these problems and evaluate their effectiveness by design and experimental assessment of prototypes.
Effective start/end date1/07/1522/11/18


  • Engineering and Physical Sciences Research Council


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.