Insect-Inspired Hexapod Robot Achieves 95% Recovery Rate in Tests

A newly developed centimeter-scale hexapod robot, INSECTRON, uses magnetic transmission to achieve robust mobility with minimal mechanical complexity. The employed magnets transmit torque, slip safely under overload, and enable contactless sensing for phase-based control, allowing stable locomotion and autonomous recovery from perturbations. These findings have been published in npj Robotics.

Closeup of an orange and white jumping spider insect navigating natural elements
Study: Insect-inspired hexapod robot with magnetically coupled compliant and modular drive system. Image Credit: LENSHAPE/Shutterstock.com

Magnetic Transmission Brings Built-In Resilience to Hexapod Robots

Legged robots draw inspiration from animals that combine mechanical redundancy, distributed control, and structural compliance to traverse cluttered terrain. Hexapod robots are particularly effective on uneven ground, offering stability, fault tolerance, and redundancy suited to search-and-rescue, planetary exploration, disaster response, and industrial inspection.

However, conventional legged robots rely on rigid joints, fixed fasteners, and precisely aligned shafts, which limit modularity, complicate servicing, and leave transmissions vulnerable to shock loads and misalignment.

Small-scale legged robots are especially susceptible to impacts, terrain irregularities, and leg entanglement that can cause mechanical failure or motor stall. Magnetic couplings have been used in fluid pumps, sealed systems, and swimming robots, but not in legged locomotion. 

INSECTRON addressed this gap by embedding magnetic couplings directly in the drivetrain, replacing conventional fasteners while providing passive overload protection and modularity.

Two Motors and Magnetic Clutches Power Six-Legged Locomotion

Locomotion is produced by two identical, magnetically coupled drive modules, one rotated 180° relative to the other, each driven by a direct-

current (DC) micromotor. Motor rotation passes through a spur-gear transmission and is converted into oscillatory leg motion by a magnetically coupled crank-driven leg mechanism, with three legs mechanically synchronized per module to produce a repeatable stance-swing gait.

The coupling uses cylindrical NdFeB N50 permanent magnets that provide a nominal holding force of 4.32 N and a remanent flux density of 1.40 T. Torque is transmitted by the attractive force between facing magnets until the transmitted torque exceeds the coupling's slip limit. At this point, the leg temporarily decouples from the drivetrain, preventing motor stall and gear damage, and re-engages automatically once the disturbance is removed.

The robot has a body length of 8.7 cm and a mass of 33.9 g. Load-bearing parts were additively manufactured in VisiJet M3 Crystal, with auxiliary components in PLA. Control electronics comprise a Seeed Studio XIAO ESP32-S3, a DRV8833 dual H-bridge, and two 49E Hall effect sensors that use the transmission magnets to reconstruct crank phase.

A phase-based controller runs at 100 Hz, selecting one module as leader at a baseline PWM duty cycle and regulating the follower with a proportional-derivative controller to maintain a 180° anti-phase relationship. 

Maneuverability was evaluated through differential turning by actuating modules at different frequencies, and spin-in-place by actuating them at equal frequency in opposite directions. Experiments were conducted in a boxed arena with 1.3 cm grid paper for motion tracking, with the robot fully untethered and telemetry streamed over Wi-Fi.

Download a free copy of this page here!

INSECTRON Recovers From Up to 95% of Controlled Disturbances

The robot integrated a compact drivetrain with passive magnetic elements that simultaneously provide overload protection, mechanical robustness, and sensing. At the gear-leg interface, magnetic couplings act as passive micro-clutches that temporarily disengage under excessive torque, protecting the gear train, preventing motor overcurrent, and re-engaging automatically once the disturbance is removed.

Additional magnetic clutches connect the two drive modules, accommodating transient load imbalances during impacts or asymmetric terrain interactions. Magnet-Hall pairs on selected gears provide contactless measurement of output gear angular position for proprioceptive feedback. 

On neutral ground, forward speed increased monotonically with drive frequency, rising from 1.8 to 2.7 cm/s, while energy per meter fell from 24 to 14 J/m and cost of transport dropped from 72 to 42. On slopes of 0°, 3°, and 5°, speed fell from 2.7 to 1.6 to 0.8 cm/s, while power remained nearly constant and energy per meter rose from 15 to 26 to 50 J/m.

Differential steering increased angular speed from 0.11 to 0.23 rad/s while turning radius decreased from 0.2 m to below 0.1 m. Spin-in-place raised angular speed from 0.71 to nearly 0.90 rad/s with modest power increases. 

Across soil, sand, perlite, and grass, stable locomotion was maintained; grass yielded the lowest cost of transport at 30, and granular perlite the highest power demand. In 360 perturbation trials that involved controlled vertical pressure applied to the robot’s body, recovery probabilities were 0.91, 0.95, and 0.94 for the three contact durations (0.1, 0.5, and one second).

Magnetic Couplings Could Make Miniature Robots More Resilient

INSECTRON demonstrates that embedding magnetic couplings directly within the drivetrain can deliver robust, controllable, and energetically competitive locomotion with only two actuators.

The magnets serve simultaneously as torque transmitters, passive micro-clutches that protect the gear train from overloads, and contactless sensing elements for phase-based control. The platform maintained stable locomotion across slopes and heterogeneous substrates, achieved predictable differential steering and spin-in-place, and autonomously recovered in 91–95% of perturbation trials. 

By eliminating mechanical fasteners and precision alignment, the architecture could enable tool-free leg replacement and rapid reconfiguration, supporting modular, resilient, and field-deployable miniature legged robots for confined and uncertain environments.

Journal Reference

Romano, D., & Manduca, G. (2026). Insect-inspired hexapod robot with magnetically coupled compliant and modular drive system. Npj Robotics, 4(1), 54. DOI:10.1038/s44182-026-00106-y. https://www.nature.com/articles/s44182-026-00106-y.

Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Nandi, Soham. (2026, November 19). Insect-Inspired Hexapod Robot Achieves 95% Recovery Rate in Tests. AZoRobotics. Retrieved on November 19, 2026 from https://www.azorobotics.com/News.aspx?newsID=16517.

  • MLA

    Nandi, Soham. "Insect-Inspired Hexapod Robot Achieves 95% Recovery Rate in Tests". AZoRobotics. 19 November 2026. <https://www.azorobotics.com/News.aspx?newsID=16517>.

  • Chicago

    Nandi, Soham. "Insect-Inspired Hexapod Robot Achieves 95% Recovery Rate in Tests". AZoRobotics. https://www.azorobotics.com/News.aspx?newsID=16517. (accessed November 19, 2026).

  • Harvard

    Nandi, Soham. 2026. Insect-Inspired Hexapod Robot Achieves 95% Recovery Rate in Tests. AZoRobotics, viewed 19 November 2026, https://www.azorobotics.com/News.aspx?newsID=16517.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.