This paper describes the development of a laparoscopic assistant robot with motor-controlled bending and zooming mechanisms with voice command motion control. The developed robot is called the Kaist Laparoscopic Assistant Robot (KaLAR). KaLAR is shown to be easy to use, compact, safe, and reliable. The voice-controlled command input is expected to enhance the overall performance of the system while reducing the control load imposed on the surgeon during a laparoscopic surgery.
Laparoscopic surgery is a type of minimally invasive surgery (MIS) in which surgery is conducted with instruments and an laparoscope inserted thorough three to five incisions (less than 10 mm in diameter) called ports in the abdomen. The abdominal region is then inflated with carbon dioxide to create a workspace in which targeted surgical site is viewed though a laparoscope equipped with a CCD camera. This type of MIS approach greatly reduces surgical trauma and pain, and shortens the length of recovery time, but the operating surgeons suffer from many limitations such as limited flexibility and range of motion, inability to ¨touch and feel〃 the organ of interest, and the lack of depth perception due to the projected image of the surgical site on a 2D monitor. Despite many restrictions, provision of a steady and focused view of the time-varying surgical site can greatly enhance the performance of the surgeon during the surgery.
For the camera control, an assistant surgeon, usually an apprentice or intern, is employed to assist the primary surgeon. However, due to the fulcrum effect, an inexperienced assistant fails to re-interpret the surgeon…s command and provides the view of undesired side. In addition, tremoring of the laparoscope-guiding hand is another issue that degrades the performance of a laparoscopic surgery. To alleviate these problems, many researchers have worked on developing assistant robots and resulted in commercialized robots such as AESOP and EndoAssist. However, these bulky robots are designed mainly to satisfy the functional requirements but are not easily applicable to the current operating room where a multitude of surgeon, anesthesiologist and nurses interact within a limited working space.
To overcome the issue in current systems, an assistant robot was designed with two major criteria: safety and adaptability. To design a robot with optimized range of motion, motions in several popular laparoscopic procedures were studied and these parameters are reflected in the design. For adaptability, the robot is designed with compact size to minimize the interference with the staffs in the operating room and is equipped with intuitive voice-activated control to minimize the training for control. Short setup time is another advantageous feature in our system.
The inherent problem of being bulky and generating wide range of motion in the current assistant robots lies in the design concept of externally generated motion using a rigid laparoscope. In order to reduce the motion range and size, I have adopted a flexible laparoscope with a bending mechanism within the abdominal cavity and a zooming mechanism through in and out motion at the port. The robot is attachable to the bedside using a conventional laparoscope holder with multiple DOF joints and is compact enough for hand-carry.
Through the result of movement feasibility test, I had founded the problem of tendon-driven mechanism. To solve this problem, I tried to reduce the hysteresis of bending section and compensate the motor-gear backlash. Applying the spring to the bending section and the compensative value to the desire trajectory, nonlinearity and hysteresis were reduced. And that was proved through the implementation of circle trajectory.
By compact design and internally generated motions within abdomen, the interference with the operating surgeon, a common problem in robot-aided surgery, is dramatically reduced. Through the comparison of outer workspace between AESOP1000 and KaLAR, it is found that the outer workspace of KaLAR is minimized. The system also is applicable in the surgery where no pivot point is produced. Experiment on test-bed shows that the developed system has the potential for clinical use upon further improvement.
