A New-Generation Soft Robot Leveraging Self-Excited Oscillation and Physical Intelligence for Colonoscopy

Overview

Colorectal cancer is one of the main causes of cancer incidence and mortality worldwide. However, traditional colonoscopy usually causes significant discomfort to patients, and due to the reliance on external passive propulsion of the equipment and sedation is often required. The existing robotic alternatives, including soft pneumatic colonoscopy robots and magnetic-driven capsule systems, although improving flexibility and safety to some extent, still face several significant limitations, such as insufficient movement efficiency, inadequate steering capability, unsatisfactory magnetic control stability, and potential compatibility issues with implanted medical devices. To address these challenges, this study proposes a new generation of soft colonoscopy robots inspired by insect movement and based on self-excited oscillation. The robot employs two self-excited oscillation units distributed circumferentially and is coupled through a fluid network, enabling synchronous rhythmic movement without the need for electronic components. By adjusting the flow rates on both sides, the system can achieve smooth linear advancement and also complete active steering through differential output forces. Parameter optimization showed that oscillation frequency increased with higher flow rate and smaller pipeline dimensions. The optimal unit size was 10mm x 80mm, reaching about 5.2Hz at 0.8L/min. The robot achieved an average forward speed of 0.418 mm/s and an average angular velocity of 2.33°/s, and successfully moved through 80-120 mm pipes even under humid conditions. Comparative tests showed that the robot moved 2.2 times faster than a traditional airbag-type robot in a straight pipe and achieved a 100% success rate in a 150° bend, compared with 40% success rate for the airbag-type robot. These results indicate that combining self-excited oscillation with physical intelligence allows this fully electromagnetic-free soft robot platform to achieve efficient axial propulsion and active steering. More importantly, this approach has the potential to make colonoscopy safer, gentler, and more comfortable for patients, reducing discomfort and lowering the need for sedation. By improving the patient experience and accessibility of colonoscopy, this technology could encourage earlier detection of colorectal cancer, expand participation in preventive screening, and ultimately generate meaningful social impact through better public health outcomes.