Ommo Technologies Raises Series A to Give Robots a 'Sixth Sense' with Magnetic Positioning
When a robotic arm reaches into a narrow cavity, fingers overlap, or it grasps a soft sponge, its biggest fear is not "can't see clearly" but "doesn't know where it is." This is the core problem Ommo Technologies is solving.
The spatial intelligence company recently completed a Series A funding round of tens of millions of dollars, co-led by Hong Kong's Dingpei Group and a well-known fund, with Kangjun Capital participating. The funds will be used for technology iteration and mass production of its core product, the "permanent magnet spatial positioning system," accelerating deployment in embodied intelligence, advanced manufacturing, and medical scenarios.
Traditional robot positioning relies on vision, encoders, force and tactile sensors. But when fingers mutually occlude, enter tight spaces, or grasp soft objects, these sensors can fail. For example, mainstream optical positioning completely loses tracking when line of sight is blocked—flexible endoscopes in minimally invasive surgery cannot work optically inside the body.
Ommo's solution, founded by Zheng Minjie, uses permanent magnet magnetic field positioning. Unlike traditional electromagnetic positioning, Ommo uses mechanical rotation of permanent magnets to generate characteristic magnetic fields, like building a small "indoor BeiDou" in physical space. A continuously rotating permanent magnet inside the host acts as a magnetic field source, while micro magnetic sensors capture magnetic signals at their location, and algorithms compute complete 6DoF position and attitude data with sub-millimeter accuracy. The sensors are as small as 0.8mm, smaller than a grain of rice, and can be embedded in robot end-effectors, wearables, or medical devices.
Ommo's permanent magnet positioning technology was first deployed in medical surgical navigation—where near-zero error tolerance makes it the best test for system precision. The company has established a quality system compliant with ISO 13485, cooperating with over a hundred medical device companies domestically and internationally, with multiple navigation devices in the regulatory approval stage.
Beyond medical scenarios, embodied intelligence is Ommo's current key focus. In robotic fine operations like grasping, inserting, and screwing, tiny position errors are easily amplified into slips or failures after contact. Meanwhile, the industry suffers from an extreme scarcity of 3D physical operation datasets. While large language models can train on vast internet text, robots' needed 3D operation data has yet to be collected at scale. Ommo's positioning system can be transferred to the embodied domain to build data collection platforms. Its first data-collection glove, equipped with medical-grade permanent magnet sensors, is used to capture human hand operation postures. Zheng admits this market is "growing much faster than I expected."
In Zheng's view, spatial intelligence is the "underlying neural center of physical AI." Large language models are the robot's brain, robotic arms are limbs, and Ommo's permanent magnet positioning system connects the real-time feedback pathway between environmental perception and limb execution, "replicating humans' innate 3D spatial perception." When robots can know "where they are" even when fingers are occluded, they move closer to truly operating as flexibly as humans.