Researchers at the National University of Singapore have developed a soft robotic system that gives machines a human-like sense of body awareness, allowing them to sense touch, movement, and external forces without relying on cameras or external tracking devices.

The breakthrough centers on proprioception, often described as the body’s “sixth sense,” which enables humans to understand the position and movement of their bodies without needing to look. Inspired by this biological ability, the research team created what they call an “expected perception” framework for soft robots.

Figure 1. Touch-Sensitive Robot.

Robots that Understand Their Own Movement

The new system enables the robot to predict how its body should move and compare those predictions with real-time sensory feedback. When the two do not match, the robot recognizes that it has come into contact with an external object or force. This approach helps solve a longstanding challenge in soft robotics: distinguishing between a robot’s own movement and interactions with its surroundings.

To demonstrate the technology, researchers equipped a flexible robot with liquid-metal-based sensors capable of detecting bending, stretching, and deformation. Using only its internal sensing system, the robot was able to navigate and respond to physical contact in real time. “Soft robots, too, need proprioception,” said Cecilia Laschi. Figure 1 shows touch-sensitive robot.

Camera-Free Navigation Through Touch

Traditional soft robots often struggle because their sensors respond both to self-generated motion and external contact, making environmental interpretation difficult. The new framework mimics how the human brain predicts sensory input. The robot calculates its expected body position based on movement commands and continuously compares those predictions with signals from its flexible sensors.

In one experiment, researchers tested the robot in a maze-navigation task where it moved autonomously without cameras. Instead, it relied entirely on touch and internal sensing to detect walls and adjust its direction.

In another test, a human operator guided the robot through movements similar to massage therapy or medical procedures performed on a manikin. The robot later reproduced those motions with high precision. “It could detect external contact within 0.4 seconds and determine the direction of applied forces with an error margin below 10 degrees, even in dynamic environments,” Laschi explained.

Potential Uses in Healthcare and Underwater Robotics

Researchers believe the technology could improve human-robot collaboration in healthcare, rehabilitation, and assistive robotics. Soft robots with advanced tactile awareness may eventually help support elderly care, assist medical professionals, or aid surgeons during minimally invasive procedures.

The system could also benefit underwater robotics, where cameras are often limited by darkness or poor visibility. Robots inspired by octopus arms, for example, could rely on touch-based sensing to navigate challenging underwater environments. The team plans to further enhance the technology using machine learning systems modeled on how the human brain develops internal representations through experience.

References:

1. https://interestingengineering.com/ai-robotics/soft-robot-sixth-sense-camera-free-navigation

Cite this article:

Keerthana S (2026), Touch-Sensitive Robot Can Navigate Like Humans Without Vision, AnaTechMaz, pp.370