A Feedforward-Moment-Gyro-Control for Positioning Wirelessly Light-Source and Wireless- Camera in Laparoscopic Instruments

This article presents a gyroscopic mechatronic system, which helps the laparoscopic surgeon to wirelessly control the zoom and panoramic position of a camera and a light source, adapted to a manipulator for minimally invasive surgery. The gyroscope adapted to the manipulator generates a reference signal used by an open loop control. The camera and light source system are mounted on an electromechanical device (robotic arm) with three degrees of freedom (3DOF). Experiments performed with the system show good pan, tilt and zoom performance of the camera and light source. Success is measured by comparing an input signal from the voltage levels generated by a transducer with micro-electro-mechanical systems (MEMS), versus the signals for the angular positions of two servo-motors (pan and tilt) and zooming in or out of the camera by a DC motor.

Abstract. -This article presents a gyroscopic mechatronic system, which helps the laparoscopic surgeon to wirelessly control the zoom and panoramic position of a camera and a light source, adapted to a manipulator for minimally invasive surgery. The gyroscope adapted to the manipulator generates a reference signal used by an open loop control. The camera and light source system are mounted on an electromechanical device (robotic arm) with three degrees of freedom (3DOF). Experiments performed with the system show good pan, tilt and zoom performance of the camera and light source. Success is measured by comparing an input signal from the voltage levels generated by a transducer with micro-electro-mechanical systems (MEMS), versus the signals for the angular positions of two servo-motors (pan and tilt) and zooming in or out of the camera by a DC motor.

Introduction
Mechatronics has actively participated in the rehabilitation of patients in about 10 % of the total world population according to the World Health Organization (WHO) [1 ̶ 3].
One of the elements widely used in medical applications is transducers [4,5], which are proposed to support surgeons in the laparoscopic field. An important topic addressed by this field is to move a laparoscopic camera and light into the abdominal cavity; the camera provides continuous video images taken from the pelvic abdominal cavity of the patient. Nowadays, the position control of camera and light is achieved by using vision, voice and mechanical interfaces. Some solutions to manipulate position in this kind of instruments came from mechatronics assistance with three degrees of freedom (PMAT) [6], where a mechanical harness is placed over the shoulders of the surgeon and provides the position for the camera view. Another system is the robotic camera assistant (EndoAssist) [7], where the surgeon moves the laparoscopic camera through a helmet equipped with an infrared light emitting diode (IR LED) transmitter, which is in contact with an LED receiver placed on a remote monitor. This system allows changing the angle of a camera placed inside the pelvic cavity of the patient.
Other methods, like those used in the automated endoscopic system for the optimal position (AESOP) and the KAIST laparoscopic assistant robot systems (KALAR), use voice commands issued by the surgeon [8,9].
Another alternative used is the single port access (SPA) [10], which consists of making a hole of 26 mm of diameter below the navel through which an independent camera is introduced and then controlled from the outside by an electromagnetic field [11].
Another minimally invasive surgical robot is the insertable robotic effector's platform (IREP) [12], which is a wire-actuated wrist with a passive flexible component arm that is introduced to manipulate the trajectory of the laparoscopic camera and light [13]. Finally, the robotic cameraman is an industrial robotic arm, whose end tip was adapted to provide video images as it moves into the abdominal cavity of a patient [14]. This paper presents a prototype mechatronic system (i.e. a feedforward-moment-gyro-control) to move wirelessly a laparoscopic wireless-camera and lightsource inside the abdominal cavity of a patient [15,16].

Principles of the system
A feedforward-control is implemented to achieve the reference position of the camera; this reference signal is given by an electronic gyroscope mounted on a laparoscopic grasper instrument. The surgeon decides when to start or stop this task by pressing with the thumb a force sensor (FSR) installed in the instrument (i.e. grasper of 5 mm of diameter).
The mechanical prototype is supported on transducer technology of micro-electro-mechanical systems (MEMS), which involves a conversion of energy into a voltage. Moreover, the output of the angular rate sensor (i.e. gyroscope) is amplified and used as a reference signal to move the camera on three different axes: pan, tilt (both are rotational) and zoom (translational). The gyroscope has a sensitivity of 300 mV/°/s as reported by their manufacturers. Some of these control applications in the medical field can be consulted in [17,18]. Concerning the issue of security related to radiate power levels for medical instruments in the operating room, there are some previous works that give some recommendations [19][20][21][22].
As shown in Figure 1, the process begins when the surgeon presses with the thumb the FSR (Mod. FSR 402). If the surgeon does not press the FSR, the laparoscopic instrument (i.e. the master manipulator) is used as an instrument of standard surgery.
When the instrument rotates, the transducer detects the angle of motion and converts it into a voltage signal command, as shown in Figure 4. Figure 3. The master manipulator was modified to insert an FSR. The surgeon can press using the thumb to start or stop the remote position of the trajectory of the laparoscopic wireless-camera.

Slave manipulator dynamic model
Given that the manipulator of 3DOF is represented by the rectilinear motion 1 ∈ ℝ and 2 angular motions 2 ∈ ℝ, 3 ∈ ℝ, where 1 stands for the zoom, 2 is the tilt displacement and 3 is the pan displacement as depicted in Figure 5, according to the modelling procedure of the Lagrange equations of motion [27]. Firstly, let us compute the kinetic energy of the manipulator, which is given by: Moreover, the manipulator is not affected by gravity, therefore the potential energy is: From eq. (1) and (2) The dynamic model of the system in joint space coordinates is as follows: Being S1 = q1 (0), S2 = q2 (0), S3 = q3 (0) ∈ ℝ, the control input τ ∈ ℝ 3 is given by the gyroscope and the strain gauge mounted on the laparoscopic ISSN: 2594-1925 grasper as is shown in Figure 3.

Feedforward control
The surgeon, when using the master manipulator instrument, would have a limited working space due to the wrist of the hand plus the small dimension of the 5-mm diameter hole of the patient. Hence, the rotation over coordinates x, y and z are restricted from 0 to 90 degrees for each axis (-α to α, -β to β, and -θ to θ respectively). Furthermore, the slave manipulator (i.e. laparoscopic wireless-camera) has a working space in spherical coordinates restricted from 0 to 90 (-φ to φ, -θ to θ) as depicted in Figure 3; the differences and restrictions are shown in Table 1A.
The relationship between the z axis in the Cartesian system and the r axis in the Spherical system is shown in Table 1B.

Master manipulator board
The gyroscope was adjusted to measure the orientation around the x-axis from -45 to 45 degrees, around the y-axis from -45 to 45 degrees and around the z-axis from 45 to -45. An algorithm was developed to position the wireless-camera and light considering protocol shown in Table 2 and Table 3.

Data acquisition system for master manipulator
To assess the matches of the position of laparoscopic wireless-camera and light versus the orientation of the gyroscope, we collected the proportional output voltage from the x, y, and z-axes of the gyroscope and converted this into a digital signal with a ISSN: 2594-1925 microcontroller [28]. The ATmega326 supports 10 bits of resolution and was adjusted to 3.3 V as a reference voltage. Therefore, 3.29 V (LSB * 1023) is, in theory, the maximum voltage available. Thus, the error can be expressed as 0.097 % (0.00323 V * 100 / 3.3 V). Figure 6 shows how the trajectory position of the servomotor is controlled. These results are representative for both angular displacements 2 and 3, but only one (pan) is reported.

Servomotor rotation degrees for pan and tilt
The servomotors that control the movement of pan and tilt properly do not have an encoder, so a mechanism was adapted to take the readings, using an incremental type DC encoder with resolution of 100 rpm at 5 V. In the same way in Figure 6 it is observed that the operating voltage of the x-axis of the gyroscope lies within the limits of the aforesaid theoretical voltage.

The zoom trajectory from DC gear motor
The data logger records the voltage signal around the z-axis. Figure 7 shows that when the orientation of the gyroscope is less than 75 degrees, the zoom is moving down; when the gyroscope is greater than 125 degrees, the zoom is moving up; and when the gyroscope rotates between 76 and 124 degrees, the zoom is stopping.

Discussion
According to the results obtained by the DAQ system to assess the position of the laparoscopic camera and light, we analyzed and displayed the error through the process of conversion by the ADC system as shown in Figure 8. For instance, for a voltage of 1.98 V in the x-axis of the gyroscope, we had a response of 45° in the servomotor. The present study showed that the position of a laparoscopic camera and light (slave manipulator) inside the abdomen of a patient can be controlled by the surgeon with a laparoscopic instrument (master manipulator). Also, showed that 300 mV/°/s of gyroscope sensitivity is enough to guide the pan and tilt view of the camera and light. Considering that at lower sensitivity we get lower resolution, we will estimate with Eq. 7 the absolute error and with Eq. 8 the scientific error.

Conclusion
Where: = ̅ ± (8) In this work, it was shown that a surgeon can modify the trajectory of a laparoscopic camera and = Sensitivity mV/°/s ̅ = Mean. n = Total Reading. X = Scientific Error.
In this way, the absolute error e = 0.20 is estimated; which represented in terms of scientific error has an error in the range of 1,089 -1,289, which represents a degree of positive sensitivity. The behavior of the experiment is shown in Figure 9. In addition, the results of the experiments show that when entering an FSR sensor to select between two different steps that represent two different spatial planes (with x and y coordinates), we can send information in the first step to the servomotors for pan and tilt motion, and the second step to send information to gear dc motor for the zoom motion.
Finally, the accuracy of the digitizing process of the output voltage for the x-axis of the gyroscope, versus the response of the electromechanical actuators (servomotors and gear motor), was demonstrated by the data record that represents the output voltage at the microcontroller AT mega328. The signal sent through the path of wireless light using a laparoscopic instrument with an embedded board containing a gyroscope. The surgeon decides when to move the laparoscopic camera in three axes (pan, tilt, and zoom). The three axes can move back and forward. Although a feedforward control system cannot correct the errors that could be generated, nor compensate the perturbations affecting the system, nonetheless, some advantages of using this type of control are its simplicity of implementation and low cost.