Despite recent advances in modern surgical robotic systems, an ongoing challenge remains their limited ability to control grasp force. This can impair surgical performance as a result of either tissue slippage or trauma from excessive grasp force. In this work we investigate a force control strategy to address this challenge based on the detection of incipient slip. Our approach employs a grasper face whose shape is engineered to encourage preferential localised slips that can be sensed using embedded displacement sensors prior to gross slip occurring. This novel approach enables closed loop control of the grasping force to prevent gross slip whilst applying minimal force. In this paper we first demonstrate the efficacy of sensing incipient slip and then demonstrate how this can form a robust closed loop grasping system to maintain stable control of tissue. Results demonstrate that this approach can achieve equivalent grasping performance to a scheme employing a fixed maximal grasping force while reducing tissue loading, and thus risk of trauma. This provides the foundation for the development of automated surgical robots with adaptive grasp force control.
The limited grasping control available in Robot Assisted Surgery is considered a significant limitation of the technology. Traditionally the integration of haptic feedback has been proposed to resolve this issue but has found limited adoption. Here we investigate an alternate approach based on the concept of detecting localised slips caused by the intrinsic elastic properties of soft tissues. This method allows for the early detection of slip so that mitigating actions can be taken before gross slip can occur, allowing the grasper to minimise the force required to maintain stable grasp control. In this paper we detail the design of a sensor developed to detect incipient slip by monitoring the relative difference in tissue movement at the front and back of the grasper, caused by tissue slip. We then demonstrate the sensor’s efficacy for the early detection of slip, as well as its ability to automate grasping under representative surgical conditions, with the automated case providing comparable performance to one which uses the maximum allowable grasp force. This work provides evidence that the slip detection methodology developed is consistently able to detect incipient slip before macro slip occurs, thus offering a strong basis for its use in automating surgical grasping tasks to avoid tissue trauma and slip.
Robotic surgical platforms have helped to improve minimally invasive surgery; however, limitations in their force feedback and force control can result in undesirable tissue trauma or tissue slip events. In this paper, we investigate a sensing method for the early detection of slip events when grasping soft tissues, which would allow surgical robots to take mitigating action to prevent tissue slip and maintain stable grasp control while minimising the applied gripping force, reducing the probability of trauma. The developed sensing concept utilises a curved grasper face to create areas of high and low normal, and thus frictional, force. In the areas of low normal force, there is a higher probability that the grasper face will slip against the tissue. If the grasper face is separated into a series of independent movable islands, then by tracking their displacement it will be possible to identify when the areas of low normal force first start to slip while the remainder of the tissue is still held securely. The system was evaluated through the simulated grasping and retraction of tissue under conditions representative of surgical practice using silicone tissue simulants and porcine liver samples. It was able to successfully detect slip before gross slip occurred with a 100% and 77% success rate for the tissue simulant and porcine liver samples, respectively. This research demonstrates the efficacy of this sensing method and the associated sensor system for detecting the occurrence of tissue slip events during surgical grasping and retraction.