# Modelling of a vibro-impact self-propelled capsule in the small intestine

## Abstract

This paper studies the modelling of a vibro-impact self-propelled capsule system in the small intestinal tract. Our studies focus on understanding the dynamic characteristics of the capsule and its performance in terms of the average speed and energy efficiency under various system and control parameters, such as capsule’s radius and length, and the frequency and magnitude of sinusoidal excitation. We find that the resistance from the small intestine will become larger once the capsule’s size or its instantaneous velocity increases. From our extensive numerical calculations, it is suggested that increasing forcing magnitude or choosing forcing frequency greater than the natural frequency of the inner mass can benefit the average speed of the capsule, and the radius of the capsule should be slightly larger than the radius of the small intestine in order to generate a reasonable resistance for capsule progression. Finally, the locomotion of the capsule along an inclined intestinal tract is tested, and the best radius and forcing magnitude of the capsule are also determined.

## Keywords

Vibro-impact Non-smooth dynamical system Self-propulsion Capsule endoscope Capsule robot## 1 Introduction

Since its introduction into clinical practice 15 years ago, capsule endoscopy (CE) has become the primary modality for examining the surface lining of the small intestine, an anatomical site previously considered to be inaccessible to clinicians. However, all the available CEs have passive locomotion systems, and their reliance on peristalsis for passage through the intestine leads to significant limitations, in particular due to their unpredictable and variable locomotion velocities. For example, intermittent high transit speeds lead to incomplete visualisation of the intestinal surface, resulting in the potential for significant abnormalities to be missed. To improve the proportion of the lining that is visualised, patients must fast for 8–12 h before the procedure and for at least 4 h after ingestion of the capsule. In most cases, they are also required to drink 1–2 L of polyethylene glycol solution 12 h before the examination, in order to clear residual intestinal contents. Furthermore, the time taken for the capsule to pass varies from 14 to 70 h, with a transit time of 2–5 h through the oesophagus and stomach, 2–6 h for the small intestine, and 10–59 h for the large intestine [1]. Reviewing the images obtained during such lengthy transit periods means that the procedure of passive CE is considered both time-consuming and burdensome for clinicians. Considering all these drawbacks, an active locomotive mechanism for CE can dramatically reduce the procedure time and allow the endoscopist to focus the examination on areas of interest. In this study, we propose the model of an on-board vibro-impact self-propelled capsule system for examining the small intestine. The aim of this study is to understand the dynamics and efficient control of the system in the small intestinal environment, so that the results presented in this paper can be utilised for prototype design and fabrication.

To design a locomotive mechanism for such a small capsule is a challenging task due to its limited on-board space. Smart materials [2] and small-scale DC motors have been used to address this issue. For example, a CE with three active flexible legs controlled by means of an on-board microcontroller was constructed by using shape memory alloy [3]. The micro-actuation concept of the shape memory alloy was also used for the development of a 6-legged endoscopic capsule [4]. Legged CEs using on-board DC motors have been designed for 4 legs [5] and 12 legs [6, 7]. A small DC motor has been included in a small capsule to simultaneously control 8 polymer treads located on the outer surface [8]. In [9], a legged capsule controlled by an on-board DC motor and a slot-follower mechanism with a lead screw was developed. Lin and Yan [10] proposed an inch-like locomotion mechanism by using DC motor-driven legs and extension/retraction of the CE body. De Falco et al. [11] reported their work of a swimming wireless capsule that utilised four propellers independently activated by DC motors. In addition to such on-board locomotive mechanisms, external magnetic fields have also been adopted for capsule propulsion, see, e.g. [12, 13, 14]. Manipulation of the external magnet can alter the locomotive direction and orientation of the capsule. Such on-board and external driving mechanisms make it feasible to either move a capsule in a limited region or anchor it at a fixed location. However, the fabrication, manipulation, system reliability, and cost of such complex devices are the main barriers of development. Our work addresses these issues by employing the so-called vibro-impact self-propulsion approach [15, 16, 17]. Its advantages over the locomotion solutions described above include the fact that all the components can be located inside the capsule and no external accessories are required. This could potentially allow for simpler sterilisation of the capsule so that make it reusable. In addition, the cost of the components needed to produce this locomotion solution is small. Together, these attributes may significantly reduce the overall cost of CE, making it an attractive proposition to healthcare providers in both developed and developing countries.

The principle of vibro-impact self-propulsion is that bidirectional rectilinear motion of the capsule can be obtained by utilising internal vibration and impact force in the presence of external resistance. A prototype of the capsule robot propelled by internal interactive force and external friction was designed by Li et al. [18], and its velocity-dependent frictional resistance inside the intestine was experimentally investigated [19]. Carta et al. [20] developed a vibrational propelled capsule composed of a motor with an eccentric mass, which can produce a reduction in the friction with the environment. The motion of a complex micro-robot exhibiting impact and friction was studied numerically and experimentally using non-smooth multibody dynamics by Nagy et al. [21]. They found that the stiction and sliding of the robot were governed by the frequency of excitation and the friction, while impact around the resonant frequency of the oscillator does not contribute to the propulsion of the robot. Numerical simulations and experimental investigations of a vibration-driven capsule system under four different friction models were studied by Wang et al. [22]. This group have also considered the planar locomotion of a vibration-driven capsule with two internal masses [23]. In the current paper, we will discuss our vibro-impact capsule, which employs additional internal impact to enhance progression [16], and analyse its dynamic characteristics in a pig small intestine. Numerical studies [24, 25] suggest that forward and backward progression of the capsule can be controlled under either fast progression or energy-saving modes in different frictional environments. Preliminary experimental studies [26, 27] have demonstrated that the vibro-impact self-propulsion technique could be a potential alternative modality for active CE, and in particular, the provision of both forward and backward progression can improve both the quality and the sensitivity of clinical examinations. It is, therefore, useful to understand how the vibro-impact self-propelled capsule can be adapted to the intestinal environment in terms of selection of system and control parameters, such as mass ratio, stiffness ratio, and frequency and amplitude of excitation for prototyping and testing.

The remainder of this paper is organised as follows: In Sect. 2, the resistances exerted on the capsule by the intestinal tract are firstly studied and then employed in mathematical modelling of the vibro-impact capsule system. In Sect. 3, bifurcation analysis is performed to study the influences of various parameters on capsule dynamics and performance in terms of average velocity and energy efficiency. Finally, some concluding remarks are drawn in Sect. 4.

## 2 Mathematical modelling

In this work, we consider the two-degrees-of-freedom dynamical capsule system as shown in Fig. 1a, where a movable internal mass \(m_1\) is driven by a harmonic excitation with forcing magnitude \(P_\mathrm{d}\) and frequency \(\Omega \). The internal mass interacts with a rigid capsule \(m_2\) via a linear spring with stiffness *k* and a viscous damper with damping coefficient *c*. The capsule has a cylindrical body with a hemispherical head and tail. Impact between the internal mass and a weightless plate connected to the capsule through a secondary spring with stiffness \(k_1\) may occur, once their relative displacement \(x_1-x_2\) is larger than or equal to the gap \(g_1\), where \(x_1\) and \(x_2\) are the absolute displacements of the internal mass and the capsule, respectively.

### 2.1 Resistances

*L*is the length of the capsule, \(R_\mathrm{c}\) is the radius of the head, the body, and the tail, \(R_\mathrm{i}\) is the original inner radius of the intestinal tract, \(\phi _\mathrm{c}\) is the angle of the point from where the intestinal tract starts to surround the capsule, and \(x_\mathrm{c}\) is the distance from the contact point to the centre of the head (or the tail).

*x*as shown in Fig. 2a, the corresponding normal pressure can be written as

*R*(

*x*) with respect to

*x*[29]. Thus, the pressure on capsule head along the x-axis can be expressed as

*g*and \(\gamma \) are the acceleration due to gravity and the inclination of the intestine, respectively.

Parameter | Value | Unit |
---|---|---|

\(R_\mathrm{i}\) | 3.9 | (mm) |

\(t_\mathrm{m}\) | 3 | (mm) |

\(E_1\) | 7 | (kPa) |

\(E_2\) | 6.3 | (kPa) |

\(E_3\) | 9.2 | (kPa) |

\(\eta _1\) | 125.9 | (kPa s) |

\(\eta _2\) | 10.3 | (kPa s) |

\(\mu \) | 0.08 | (–) |

Physical parameters of the pig small intestinal tract used in the following simulations are listed in Table 1. Based on these parameters, calculations of the resistances due to hoop stress and the gravity of the capsule with respect to capsule velocity are presented in Fig. 3, where \(F_\mathrm{hoop}\) in Fig. 3a is a typical Stribeck friction and \(F_\mathrm{gravity}\) in Fig. 3b is a classical Coulomb friction. Figure 3c, d illustrates variations of the resistance caused by the hoop stress \(F_\mathrm{hoop}\) in terms of capsule’s radius \(R_\mathrm{c}\) and length *L*, respectively. From these two figures, one can observe that both the threshold and the maximal value of the resistance will increase if either \(R_\mathrm{c}\) or *L* increases, and the resistance \(F_\mathrm{hoop}\) is more sensitive to capsule’s radius \(R_\mathrm{c}\).

### 2.2 Equations of motion

## 3 Bifurcation analysis

As described by Eq. (17), a periodic driving force is implemented to overcome the environmental resistance for capsule progression. Intuitive speaking, a large driving force and a small resistance are preferred for fast capsule progression. As can be seen from Fig. 3, resistance becomes larger once the radius or the length of the capsule increases. Therefore, our bifurcation analysis in this paper will focus on the driving force and the dimension of the capsule system. For bifurcation diagrams, we have adopted the velocity \(v^*_1\)–\(v^*_2\), which is a projection of the Poincaré map on the \(v_1\)–\(v_2\) axis, as a function of the magnitude/frequency of the driving force. Calculations were performed for 300 cycles of the driving force, and the data for the first 280 cycles were omitted to ensure steady-state responses, whereas the next 20 values of the relative velocity, \(v^*_1\)–\(v^*_2\), were plotted in bifurcation diagrams.

*N*and \(T=\frac{2\pi }{\Omega }\) are the number of cycles and the period of driving force, respectively. For simplicity, the abbreviation P-

*m*-

*n*is used to denote the period-

*m*motion with

*n*impacts per period of the driving force.

### 3.1 Influence of intestinal resistances

As the radius of the capsule, \(R_\mathrm{c}\), increases further, as shown in Fig. 6, the responses of the capsule become simpler compared to the responses obtained for \(R_\mathrm{c}=4\) (mm). As can be seen from Fig. 6a, b, both cases are dominated by period-two motion and the time of impact per period of excitation switches from three to two via grazing bifurcation. To illustrate the coexistence of two different attractors observed in Fig. 6, the region for \(P_\mathrm{d}\in [6.8,9]\) in Fig. 6b was marked by netted blue to indicate the coexisting P-2-3 and P-2-2 motions. Their corresponding basins of attraction at \(P_\mathrm{d}=7\) (mN), 8 (mN), and 9 (mN) are presented in Fig. 6e–g, respectively, where the initial displacement and velocity of the capsule were fixed as zero, with only the initial conditions of \(x_1\) and \(v_1\) varying. As can be seen from the basins, there are two white regions for the initial conditions leading to P-2-3 motion with all the other purple region resulting in P-2-2 motion. As the forcing magnitude increases, the basin of P-2-3 shrinks slightly.

*L*, has a significant influence on the resistance in the intestinal tract. Intuitively, the longer of the capsule, the larger resistance in the tract, so that the slower of capsule progression. Figure 7 presents the average speeds and the energy efficiencies of the capsule for various lengths of the capsule, and different periodic responses of the capsule are marked in the figure. As can be seen from Fig. 7a, period doubling is critical for capsule progression, since the period-one motions produced by small driving force cannot overcome the resistances in the tract, but visible progression can be observed after the period-doubling bifurcation from P-1-1 to P-2-2. As \(P_\mathrm{d}\) increases, the P-2-2 motion on each curve successively bifurcates into P-2-3 via grazing bifurcation, except for the red curve for \(L=11\) (mm). In addition, the green and purple curves for long capsules (\(L=14\) (mm) and 17 (mm)) undergo another grazing bifurcation when the driving force is sufficiently large, which switches P-2-3 into P-2-2 again. Comparing both \(v_\mathrm{avg}\) and \(x_E\), our calculations prove that a shorter capsule has faster average speed and better efficiency for energy consumption.

To sum up, it can conclude that increasing the driving force will benefit the average velocity of the capsule but can decrease its corresponding energy efficiency. It has also shown that increasing the capsule’s size will enlarge the resistances on the capsule introducing more period-two responses for the system. The occurrence of period-two motion always slows down the capsule and decreases its energy efficiency. However, this does not mean that the capsule’s size needs to be as small as possible, because larger resistances can restrain capsule’s backward motion and produce faster progression when the driving force is sufficiently large.

### 3.2 Influence of the magnitude and frequency of the driving force

### 3.3 Influence of the natural frequency of the inner mass

*k*has larger average velocity. Regardless of the stiffness of the primary spring, the fastest progression was achieved just after the occurrence of the grazing bifurcation when the capsule bifurcates from P-1-3 to P-1-2 motion. Figure 10d shows that the efficiencies of the capsule for different stiffness are very close, so changing the stiffness of the primary spring does not affect the efficiency of the system.

Apart from the stiffness of the primary spring, the weight of the inner mass, \(m_1\), also affects the natural frequency of the inner mass. When \(m_1\) is increased from 0.001 (kg) to 0.003 (kg), as shown in Fig. 11, the number of period-doubling bifurcations reduces. Since period-two motion retards capsule velocity, it can be observed from Fig. 11d that the capsule has a faster average velocity when its inner mass is heavier. However, the efficiency of the capsule is not affected by \(m_1\) as their local maxima are very close as shown in Fig. 11e.

### 3.4 Influence of the stiffness of the secondary spring

The stiffness of the secondary spring is another control parameter affecting the performance of the capsule. As shown in Fig. 12, hardening the secondary spring enlarges the parametric region of period-two motion degrading the average speed of the capsule. For \(k_1=4\) (N/m), there is only a small region for P-2-2 motion. When \(k_1\) was increased to 8 (N/m), the region was expanded and a grazing bifurcation for the switching from P-2-3 to P-2-2 was recorded. When \(k_1=12\) (N/m), an additional period doubling from P-2-2 to P-4-6 was observed. Comparing the average velocity of the capsule, as can be seen from Fig. 12d, the capsule with a stiffer secondary spring has faster average speed when the driving frequency is low (i.e. before the grazing bifurcation from P-1-2 to P-1-1), but slower average speed for high driving frequency (i.e. after the grazing bifurcation). Similar trend can be observed from the energy efficiency presented in Fig. 12e. It can be seen that, after the grazing bifurcation from P-1-2 to P-1-1, the energy efficiencies of the capsules with different secondary springs are similar, but the period-doubling bifurcation degrades the performance of the capsule. After the reverse period doubling, i.e. when the dynamics of the capsule bifurcates into P-1-1 motion, the capsule with a stiffer secondary spring has better energy efficiency.

### 3.5 Influence of the contact gap

The influence of the contact gap between the inner mass and the secondary spring is considered in this subsection. Firstly, a negative gap, \(g_1=-3\) (mm), representing a prestressed secondary spring, is studied, and its bifurcation diagram as a function of the forcing frequency, \(\Omega \), is presented in Fig. 13. The bifurcation diagram shown in Fig. 13a has similar bifurcation pattern as the previous ones (i.e. a grazing bifurcation followed by a period doubling), but around \(\Omega =25\) (rad/s), which is slightly higher than the natural frequency of the inner mass, the capsule has chaotic and period-3 motions.

When the driving frequency is low, \(\Omega <12.96\) (rad/s), the capsule experiences P-1-4 motion as demonstrated in Fig. 13d, m, and bifurcates into P-1-3 via a grazing bifurcation. The capsule bifurcates from P-1-3 to P-1-2 via the second grazing bifurcation recorded at \(\Omega =16.66\) (rad/s) with the coexistence of P-1-3 and P-1-2 motions recorded for \(\Omega \in [16.26, 16.66]\) (rad/s) and then from P-1-2 to P-2-4 through a period doubling at \(\Omega =16.86\) (rad/s). For \(\Omega =18.26\) (rad/s), a reverse period doubling occurs and the capsule experiences P-1-2 again until \(\Omega =23.86\) (rad/s), where a cascade of period-doubling bifurcations leads the capsule to a chaotic motion for \(\Omega \in [24.26,25.96]\) (rad/s) including a small window of P-3-5 motion for \(\Omega \in [24.46,25.06]\) (rad/s). As the frequency increases, a cascade of reverse period-doubling bifurcations was recorded, and the capsule eventually settles down at a P-1-1 motion.

The maximum average speed of the capsule can be obtained at where the grazing bifurcation from P-1-3 to P-1-2 occurs as shown in Fig. 13b. As the frequency increases, the average speed of the capsule decreases. When the frequency was increased to \(\Omega =\omega _\mathrm{n}=22.36\) (rad/s), capsule speed was enhanced again by the resonance. As can be observed from Fig. 13c, the P-1-1 motion after the reverse period doubling has the best efficiency in energy consumption. However, the energy efficiency of the P-1-2 motion (for the maximum average speed) is not far from the best efficiency obtained by the P-1-1 motion, so the P-1-2 motion is a better choice in terms of both performance indices.

Bifurcation diagrams for different gaps under variation of the forcing frequency are shown in Fig. 14. As the gap increases, bifurcations of the capsule become simpler. For example, as can be seen from Fig. 14a, the regions for chaotic motion and coexisting P-1-3 and P-1-2 motions were significantly shrunk compared to the one for \(g_1=-3\) (mm). When the gap becomes positive as shown in Fig. 14b, c, chaotic motions are completely removed and only period-two motions exist. Comparing the average velocity of the capsule shown in Fig. 14d, it can be seen that local maxima of the average velocities for different gaps can be obtained after each grazing bifurcation, and the capsule with \(g_1=3\) (mm) has the maximal average velocity after its grazing bifurcation from P-1-2 to P-1-1. Energy efficiencies presented in Fig. 14e demonstrate that the capsules with different gaps have similar efficiencies for energy consumption after their grazing bifurcations from P-1-2 to P-1-1.

### 3.6 Progression in an inclined intestine

Our previous studies have focused on the capsule progression along a horizontal small intestinal tract. In the real environment, as the gastrointestinal tract is folded inside human body, it may require the capsule to progress along an inclined intestine, i.e. \(\gamma >0\). Influence of the inclined slope on capsule progression is studied here by calculating the average velocity of the capsule as a function of the forcing magnitude as presented in Fig. 15. It can be seen that the capsule has zero average velocity for all the inclined slopes when driving force is small. As the magnitude of the driving force increases, forward progression of the capsule can be observed, where the capsule moves faster for smaller angle of the slope. It is worth noting that a larger driving force does not always lead to a faster progression, especially when the angle of the slope is large, e.g. \(\gamma =0.25\) (rad) and 0.35 (rad), which yields backward progression (negative average velocity of the capsule) as the driving force increases.

To progress forward along an inclined intestine, the capsule needs to overcome its own gravity, and the only external resource it can utilise is the environmental resistance. According to Fig. 3c, the capsule with a larger radius is required to generate larger resistance so that the capsule can progress forward along a steeper slope. This study is shown in Fig. 16, where the average velocity of the capsule was calculated as a function of the forcing magnitude under variations of capsule’s radius. As can be seen from Fig. 16a, when the forcing magnitude increases, the average speed of the capsule starts to successively increase from zero to a positive value. However, the capsule with \(R_\mathrm{c}=4.1\) (mm) firstly slows down its forward progression and begins to move backward when \(P_\mathrm{d}=28.4\) (mN). When \(R_\mathrm{c}\) is 4.2 (mm), the capsule moves slower than that for \(R_\mathrm{c}=4.1\) (mm) when \(P_\mathrm{d}<13.9\) (mN), but has a better progression when \(P_\mathrm{d}\ge 13.9\) (mN). Similar phenomenon can be observed when \(R_\mathrm{c}\) is increased to 4.3 (mm), which yields a faster progression than that for \(R_\mathrm{c}=4.2\) (mm) once \(P_\mathrm{d}\ge 29.2\) (mN). Based on our calculations, the capsule with \(R_\mathrm{c}=4.2\) (mm) and the forcing magnitude, \(P_\mathrm{d}\in [13.9, 29.2)\) (mN), is the reasonable dimension and operational regime for capsule design and control.

## 4 Conclusions and future work

In this paper, we studied the modelling of a vibro-impact self-propelled capsule system moving in the small intestine. Our studies focused on exploring the dynamics of the system and its performance in terms of the average velocity and energy efficiency under various system and control parameters, such as the forcing frequency and magnitude of excitation, the natural frequency of the inner mass, the contact gap between the inner mass and the secondary spring, and the capsule’s radius and length. We also considered the capsule’s progression along an inclined intestine and its optimum design and control parameters.

Under the assumption that the intestinal tract fully contacts with capsule surface, the intestinal resistance exerted on the capsule can be modelled using hoop strain and stress. It was found that the resistance and its threshold become larger when the capsule’s size or its instantaneous velocity increases. We also found that strengthening the forcing magnitude of excitation can benefit the average velocity of the capsule, but will lead to low energy efficiency. Increasing the radius and length of the capsule could result in resistance enhancement, which can simplify its bifurcation pattern, enlarge the parametric regime of period-two motion, and decrease capsule’s average velocity and energy efficiency. However, if the magnitude of the driving force is sufficiently large, the capsule having a larger resistance can achieve a faster forward progression.

Our investigation on the natural frequency of the inner mass shows that, when the driving frequency is relatively lower than its natural frequency, successive grazing bifurcations will occur, and this will decrease the times of impact at each period but drastically increase the average velocity of the capsule. When the forcing frequency is chosen to be in the vicinity of the natural frequency, period doubling can be observed, which leads to sudden drops of average velocity and energy efficiency. As the forcing magnitude is increased, average velocity will be decreased at low forcing frequencies while be increased at high forcing frequencies (i.e. the frequency greater than the natural frequency). Our calculations also reveal that reducing the natural frequency of the inner mass can improve capsule’s average velocity. However, this will not affect the energy efficiency of the system.

The stiffness of the secondary spring and the contact gap between the inner mass and the secondary spring was studied under variation of forcing frequency. For a stiffer secondary spring, the capsule has a faster average velocity under a low driving frequency, while a slower progression when the driving frequency becomes high (i.e. the frequency greater than its natural frequency). The dynamics of the capsule is complicated under a prestressed secondary spring, leading to a small window of chaos and period-three motion when the forcing frequency is a branching parameter. When the gap is increased, the parametric regime of period-one motion can be enlarged, which can simplify the dynamics of the capsule. Our studies also indicate that the local maxima of the average velocities for different gaps can be obtained after each grazing bifurcation, and the capsule with \(g_1=3\) (mm) has the maximal average speed after its grazing bifurcation from period-one with two impacts to period-one motion with one impact. Furthermore, varying the contact gap cannot improve the energy efficiency of the system.

Our final study focused on the locomotion of the capsule along an inclined intestinal tract. As the magnitude of the driving force increases, the capsule can move forward on a slope with an inclined angle up to \(\gamma =0.35\) (rad). However, larger magnitude of the driving force will not help capsule’s forward progression, especially for a steeper slope, e.g. \(\gamma =0.35\) (rad). It was found that, along a steeper slope, the capsule always has a slower velocity due to gravity and insufficient resistance. To overcome gravity, the capsule with \(R_\mathrm{c}=4.2\) (mm) and the forcing magnitude, \(P_\mathrm{d}\in [13.9, 29.2)\) (mN), is a reasonable choice for locomotion control.

In conclusion, our numerical studies based on a pig small intestine with the radius of \(R_\mathrm{i}=3.9\) (mm) suggest the following optimum design and control parameters as a design guideline, capsule’s radius \(R_\mathrm{c}=4.2\) (mm) and length \(L=10\) (mm), forcing frequency \(\Omega >30\) (rad/s) and magnitude \(P_\mathrm{d}>15\) (mN), natural frequency of the inner mass \(\omega _n<25\) (rad/s), stiffness of the secondary spring \(k_1=4\) (N/m), and the gap between the inner mass and the secondary stiffness \(g_1=3\) (mm).

Future works include prototype design and fabrication, test rig design, and experimental testing of the capsule prototype. Design and fabrication of the capsule prototype will be based on the numerical studies in this paper, and an artificial intestinal environment will be built for experimental testing of the prototype. Research findings along this direction will be reported in a separate publication in due course.

## Notes

### Acknowledgements

This work has been supported by EPSRC under Grant No. EP/P023983/1, and was partially supported by the National Natural Science Foundation of China (Grant No. 11672257). Dr Yao Yan’s work has been supported by the National Natural Science Foundation of China (Grant No. 11872147) and the R&D Program for International S&T Cooperation and Exchanges of Sichuan province (Grant No. 2018HH0101).

### Compliance with ethical standards

### Conflict of interest

The authors declare that they have no conflict of interest concerning the publication of this manuscript.

## References

- 1.Lee, Y.Y., Erdogan, A., Rao, S.S.: How to assess regional and whole gut transit time with wireless motility capsule. J. Neurogastroenterol. Motil.
**20**(2), 265–270 (2014)CrossRefGoogle Scholar - 2.Manfredi, L., Huan, Y., Cuschieri, A.: Low power consumption mini rotary actuator with SMA wires. Smart Mater. Struct.
**26**(11), 115003 (2017)CrossRefGoogle Scholar - 3.Tognarelli, S., Quaglia, C., Valdastri, P., Susilo, E., Menciassi, A., Dario, P.: Innovative stopping mechanism for esophageal wireless capsular endoscopy. Procedia Chem.
**1**(1), 485–488 (2009)CrossRefGoogle Scholar - 4.Gorini, S., Quirini, M., Menciassi, A., Pernorio, G., Stefanini, C., Dario, P.: A novel SMA-based actuator for a legged endoscopic capsule. In: The First IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics, 2006. BioRob 2006, pp. 443–449. IEEE (2006)Google Scholar
- 5.Quirini, M., Menciassi, A., Scapellato, S., Stefanini, C., Dario, P.: Design and fabrication of a motor legged capsule for the active exploration of the gastrointestinal tract. IEEE/ASME Trans. Mechatron.
**13**(2), 169–179 (2008)CrossRefGoogle Scholar - 6.Quaglia, C., Buselli, E., Webster III, R.J., Valdastri, P., Menciassi, A., Dario, P.: An endoscopic capsule robot: a meso-scale engineering case study. J. Micromech. Microeng.
**19**(10), 105007 (2009)CrossRefGoogle Scholar - 7.Valdastri, P., Webster III, R.J., Quaglia, C., Quirini, M., Menciassi, A., Dario, P.: A new mechanism for mesoscale legged locomotion in compliant tubular environments. IEEE Trans. Robot.
**25**(5), 1047–1057 (2009)CrossRefGoogle Scholar - 8.Sliker, L.J., Kern, M.D., Schoen, J.A., Rentschler, M.E.: Surgical evaluation of a novel tethered robotic capsule endoscope using micro-patterned treads. Surg. Endosc.
**26**(10), 2862–2869 (2012)CrossRefGoogle Scholar - 9.Zahn, J.D., Talbot, N.H., Liepmann, D., Pisano, A.P.: Microfabricated polysilicon microneedles for minimally invasive biomedical devices. Biomed. Microdev.
**2**(4), 295–303 (2000)CrossRefGoogle Scholar - 10.Lin, W., Yan, G.: A study on anchoring ability of three-leg micro intestinal robot. Engineering
**4**(8), 477–483 (2012)CrossRefGoogle Scholar - 11.De Falco, I., Tortora, G., Dario, P., Menciassi, A.: An integrated system for wireless capsule endoscopy in a liquid-distended stomach. IEEE Trans. Biomed. Eng.
**61**(3), 794–804 (2014)CrossRefGoogle Scholar - 12.Keller, J., Fibbe, C., Volke, F., Gerber, J., Mosse, A.C., Reimann-Zawadzki, M., Rabinovitz, E., Layer, P., Swain, P.: Remote magnetic control of a wireless capsule endoscope in the esophagus is safe and feasible: results of a randomized, clinical trial in healthy volunteers. Gastrointest. Endosc.
**72**(5), 941–946 (2010)CrossRefGoogle Scholar - 13.Swain, P., Mosse, C.A., Volke, F., Gerber, J., Keller, J.: 511i: in vivo studies of the potential and limitations of remote control of functional wireless capsule endoscopes with rare earth magnetic inclusions in an extra-corporeal magnetic field. Gastrointest. Endosc.
**71**(5), AB123 (2010)CrossRefGoogle Scholar - 14.Carpi, F., Kastelein, N., Talcott, M., Pappone, C.: Magnetically controllable gastrointestinal steering of video capsules. IEEE Trans. Biomed. Eng.
**58**(2), 231–234 (2011)CrossRefGoogle Scholar - 15.Chernousko, F.L.: The optimum rectilinear motion of a two-mass system. J. Appl. Math. Mech.
**66**, 1–7 (2002)MathSciNetCrossRefGoogle Scholar - 16.Liu, Y., Wiercigroch, M., Pavlovskaia, E., Yu, H.: Modelling of a vibro-impact capsule system. Int. J. Mech. Sci.
**66**, 2–11 (2013)CrossRefGoogle Scholar - 17.Jiang, Z., Xu, J.: Analysis of worm-like locomotion driven by the sine-squared strain wave in a linear viscous medium. Mech. Res. Commun.
**85**, 33–44 (2017)CrossRefGoogle Scholar - 18.Li, H., Furuta, K., Chernousko, F.L.: Motion generation of the capsubot using internal force and static friction. In: Proceedings of the 45th IEEE Conference on Decision and Control, San Diego, CA, USA, pp. 6575–6580 (2006)Google Scholar
- 19.Zhang, C., Liu, H., Tan, R., Li, H.: Modeling of velocity-dependent frictional resistance of a capsule robot inside an intestine. Tribol. Lett.
**47**, 295–301 (2012)CrossRefGoogle Scholar - 20.Carta, R., Sfakiotakis, M., Pateromichelakis, N., Thoné, J., Tsakiris, D., Puers, R.: A multi-coil inductive powering system for an endoscopic capsule with vibratory actuation. Sens. Actuators A Phys.
**172**(1), 253–258 (2011)CrossRefGoogle Scholar - 21.Nagy, Z., Leine, R.I., Frutiger, D.R., Glocker, C., Nelson, B.J.: Modeling the motion of microrobots on surfaces using nonsmooth multibody dynamics. IEEE Trans. Robot.
**28**, 1058–1068 (2012)CrossRefGoogle Scholar - 22.Wang, S., Zhan, X., Xu, J.: Comparative analysis of friction in vibration-driven system and its dynamic behaviors. J. Adv. Appl. Math.
**2**, 23–42 (2017)CrossRefGoogle Scholar - 23.Zhan, X., Xu, J., Fang, H.: Planar locomotion of a vibration-driven system with two internal masses. Appl. Math. Model.
**40**, 871–885 (2016)MathSciNetCrossRefGoogle Scholar - 24.Liu, Y., Pavlovskaia, E., Wiercigroch, M.: Vibro-impact responses of capsule system with various friction models. Int. J. Mech. Sci.
**72**, 39–54 (2013)CrossRefGoogle Scholar - 25.Liu, Y., Wiercigroch, M., Pavlovskaia, E., Peng, Z.K.: Forward and backward motion control of a vibro-impact capsule system. Int. J. Nonlinear Mech.
**70**, 30–46 (2015)CrossRefGoogle Scholar - 26.Liu, Y., Pavlovskaia, E., Wiercigroch, M.: Experimental verification of the vibro-impact capsule model. Nonlinear Dyn.
**83**, 1029–1041 (2016)CrossRefGoogle Scholar - 27.Yan, Y., Liu, Y., Páez Chávez, J., Zonta, F., Yusupov, A.: Proof-of-concept prototype development of the self-propelled capsule system for pipeline inspection. Meccanica
**53**(8), 1997–2012 (2017)MathSciNetCrossRefGoogle Scholar - 28.Kim, J.S., Sung, I.H., Kim, Y.T., Kwon, E.Y., Kim, D.E., Jang, Y.H.: Experimental investigation of frictional and viscoelastic properties of intestine for microendoscope application. Tribol. Lett.
**22**, 143–149 (2006)CrossRefGoogle Scholar - 29.Kim, J.-S., Sung, I.-H., Kim, Y.-T., Kim, D.-E., Jang, Y.-H.: Analytical model development for the prediction of the frictional resistance of a capsule endoscope inside an intestine. Proc. Inst. Mech. Eng. H
**221**(8), 837–845 (2007)CrossRefGoogle Scholar

## Copyright information

**OpenAccess**This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.