Upper gastrointestinal (UGI) tract pathology is common worldwide. With recent advancements in robotics, innovative diagnostic and treatment devices have been developed and several translational attempts made. This review paper aims to provide a highly pictorial critical review of robotic gastroscopes, so that clinicians and researchers can obtain a swift and comprehensive overview of key technologies and challenges. Therefore, the paper presents robotic gastroscopes, either commercial or at a progressed technology readiness level. Among them, we show tethered and wireless gastroscopes, as well as devices aimed for UGI surgery. The technological features of these instruments, as well as their clinical adoption and performance, are described and compared. Although the existing endoscopic devices have thus far provided substantial improvements in the effectiveness of diagnosis and treatment, there are certain aspects that represent unwavering predicaments of the current gastroenterology practice. A detailed list includes difficulties and risks, such as transmission of communicable diseases (e.g., COVID-19) due to the doctor–patient proximity, unchanged learning curves, variable detection rates, procedure-related adverse events, endoscopists’ and nurses’ burnouts, limited human and/or material resources, and patients’ preferences to choose non-invasive options that further interfere with the successful implementation and adoption of routine screening. The combination of robotics and artificial intelligence, as well as remote telehealth endoscopy services, are also discussed, as viable solutions to improve existing platforms for diagnosis and treatment are emerging.

Soft robots have been leveraged in considerable areas like surgery, rehabilitation, and bionics due to their softness, flexibility, and safety. However, it is challenging to produce two same soft robots even with the same mold and manufacturing process owing to the complexity of soft materials. Meanwhile, widespread usage of a system requires the ability to replace inner components without highly affecting system performance, which is interchangeability. Due to the necessity of this property, a hybrid adaptive controller is introduced to achieve interchangeability from the perspective of control approaches. This method utilizes an offline-trained recurrent neural network controller to cope with the nonlinear and delayed response from soft robots. Furthermore, an online optimizing kinematics controller is applied to decrease the error caused by the above neural network controller. Soft pneumatic robots with different deformation properties but the same mold have been included for validation experiments. In the experiments, the systems with different actuation configurations and the different robots follow the desired trajectory with errors of $\mathbf{3.3\pm 2.9\%}$ and $\mathbf{4.3\pm 4.1\%}$ compared with the working space length, respectively. Such an adaptive controller also shows good performance on different control frequencies and desired velocities. This controller is also compared with a model-based controller in simulation. This controller endows soft robots with the potential for wide application, and future work may include different offline and online controllers. A weight parameter adjusting strategy may also be proposed in the future.

Soft-tethered colonoscopes were proposed for safe and effective colon navigation, yet the deployment of front-wheel actuated colonoscopes is hindered by contact interactions with the lumen along the entire soft tether. To mitigate this problem, this study introduces an over-the-tube flexible device aimed to assist colonoscope deployment. The device is composed of three pneumatically driven actuators devised to repeatedly perform a two-phase operation: (phase I) to advance along the tether up to a working position relatively close to the colonoscope's tip; (phase II) to clamp and drag the tether forward, upon anchoring to the colonic wall. This way, a distal tether portion is freed, thus reducing the aforementioned limitations and fostering effective front-wheel navigation. Considering anatomical/clinical constraints and a 2N resistive force, we designed and prototyped a system with an inner and outer diameter of 12 and 26 mm, respectively, a length of 91 mm, and operating pressures equal to 150, 50, and 15 kPa for clamping the tether, elongating the device and safely anchoring to the colonic wall, respectively. The device was successfully tested, achieving locomotion speeds up to 4.9 and 2.2 mm/s, and tether-freeing rates up to 2.9 and 1.8 mm/s, in tabletop conditions and in a colon phantom, respectively.

Soft crawling robots demonstrated high compliance and effectiveness in performing complex tasks in unstructured and harsh environments. They can navigate inside constrained spaces and provide superior adaptability. This paper presents a soft crawling robot with a modified Yoshimura origami-based central chamber (elongation/contraction actuator) and four electrostatic adhesion feet (anchoring elements). It was designed to perform linear and steering locomotion under specific actuation sequences to avoid obstacles autonomously, and it features a height-adjustable ability to squeeze under low gaps. A dual-morphing mechanism, enabling the origami-based chamber to have two corresponding locomotion modalities, was thus investigated to provide a simple but effective actuation method. Tests were carried out to validate the dual-morphing mechanism and to characterise the crawling robot’s performance. Experimental tests successfully demonstrated the robot’s capabilities, e.g. locomotion under low gaps (i.e. 20 mm, 66% of the height of the robot), obstacle avoidance, climbing on a sloped surface (i.e. 15 deg), and lifting and carrying objects (i.e. 80 g, ten times its weight).

The nonlinearity and hysteresis of soft robot motions present challenges for control. To solve these issues, the Jacobian controller has been applied to approximate the nonlinear behaviors in a linear format. Accurate controllers like neural networks (NNs) can handle delayed and nonlinear motions, but they require large datasets and exhibit low adaptability. Based on a novel analysis on these controllers, we propose an adaptive extended Jacobian controller, AdapJ, for soft manipulators. This controller retains the concise format of the Jacobian controller but introduces independent parameters. Similar to NNs, its initialization and updating mechanism leverages the inverse model without building the corresponding forward model. In the experiments, we first compare the performance of the Jacobian controller, model predictive controller (MPC), NN controller, iterative feedback controller (IFC), and AdapJ in simulation. We further analyze how AdapJ parameters adapt in response to the physical property change. Then, real-world experiments have validated that AdapJ outperforms the NN controller, MPC, and IFC with fewer training samples and adapts robustly to varying conditions, including different control frequencies, material softness, and external disturbances. Future work may include online adjustment of the controller format and adaptability validation in more scenarios.