Adaptive Self Learning Robotic Linishing and Polishing

点击翻译按钮获取中文摘要

_Variability_ _is traditionally the enemy of automation__, but with today's focus on customisation and choice this needs to change. To be widely used and adopted across UK manufacturing. automation must be able to handle large dimensional changes and be capable of changing quickly and effectively between different operations. This project addresses this issue with the development and application of an adaptive path planning system that will alter the pre-set robot program whilst in use so that changes in part shape and size can be accommodated.__A technology demonstrator of an automated linishing and polishing_ _automation cell will be built_ _to highly polish plastic parts. The system will have the capability to measure and react to the size, shape and requirements of the part to be processed thereby making it self-programming and adapting.__Blanson_ _are a manufacturer of highly engineered acrylic parts and will be supported by two SME, robotic technology partners, A3L and HAL Robotics. A3L are robotic integrators focused on delivering automated finishing operations to SME's but they hope to become a high-end global supplier of automated finishing solutions. HAL Robotics are specialists in adaptive robot path planning, their software Framework allows the use of sensor or/and measurement data to alter robot programs 'on the fly' thereby allowing for the process to be flexible and controlled by the attributes of the part._ A3L will work together with HAL to deliver an innovative finishing solution that can be utilised within Blanson's manufacturing facility. We are aiming for a 'one size fits all' approach to finishing automation, i.e., the system will be able to process multiple parts without reprogramming. It will utilise sensor data to determine where the process needs to be applied and when it is complete i.e., surface roughness or part geometry readings will be fed back into a software solution which will determine fitness and automatically generate any necessary corrective robot toolpaths. AI will also be used to process data to optimise use of abrasive consumables. Current best practice is to use the abrasives for a set time period and then dispose of them (whether they are spent or not), we believe the volume of consumables used monthly can be greatly reduced delivering huge cost savings and driving a greener approach.