IIT Gandhinagar Researchers Develop Novel Control Framework for Flexible Continuum Robots to Navigate Tight Spaces

IIT Gandhinagar researchers develop the VAS framework to give flexible robots surgical-grade precision in tight spaces, revolutionizing automation and medicine.

By: AXL Media

Published: Apr 1, 2026, 4:50 AM EDT

Source: Information for this report was sourced from Indian Institute of Technology Gandhinagar

The Breakthrough in Flexible Robotic Navigation

Researchers at the Indian Institute of Technology Gandhinagar (IITGN) have introduced a transformative control framework designed to enhance the utility of continuum robots (CRs) in highly restricted environments. Unlike traditional rigid-arm robots that rely on fixed joints and stiff mechanical movements, continuum robots utilize flexible bodies to maneuver through cluttered or delicate spaces. This development is particularly significant for fields such as surgical robotics, where navigating the intricate nooks and crannies of the human body requires a level of dexterity that rigid machines simply cannot provide without risking tissue damage.

The Mechanics of Tendon Driven Systems

The specific focus of the IITGN study involves tendon-driven continuum robots (TDCRs), which mimic the fluid movements of biological appendages like an elephant’s trunk or an octopus’s tentacle. These machines utilize thin internal wires, known as tendons, attached along a flexible backbone to facilitate smooth bending in multiple directions. While these robots are cost-effective and lightweight, they are notoriously difficult to control due to their infinite degrees of freedom. According to Dr. Madhu Vadali, an Associate Professor at IITGN, the challenge lies in the fact that moving one section often inadvertently affects another, creating a complex puzzle for operators attempting to achieve precise shapes.

Simplifying Control via Virtual Actuation Space

To overcome these traditional control hurdles, the research team developed the concept of Virtual Actuation Space (VAS). This framework acts as a simplified digital layer that imagines the robot’s motion rather than calculating the physical tension of every individual tendon. By representing a section's bending through only two parameters, direction and magnitude, the system bypasses heavy computational requirements that usually hinder real-time performance. This independent control allows for multi-section robots to move more fluidly, ensuring that the movement of the robot's base does not disrupt the delicate positioning of its tip.