Insider Brief
- LG Innotek and TDK are partnering to develop sensing modules for physical AI systems, including robot vision and tactile sensing.
- The companies said the work will combine LG Innotek’s optical and precision module-design capabilities with TDK’s sensor technologies, with a vision module using an inertial sensor planned for 2027.
- LG Innotek said the partnership will also focus on tactile sensing, real-time processing and sensor-data analysis as robot makers look to combine more sensing functions into single systems.
LG Innotek and TDK are teaming up to develop sensing modules for physical AI systems, including robot vision and touch.
The companies plan to combine LG Innotek’s optical and precision module-design capabilities with TDK’s sensor technologies. The work is expected to focus on next-generation vision sensing modules and tactile sensing modules for robots, according to the companies.
“By combining TDK’s diverse fundamental technologies and sensors with LG Innotek’s core capabilities, we will add value for customers and bring truly revolutionary products to the global market,” TDK president and CEO Noboru Saito said in the announcement.
LG Innotek said the partnership is aimed at building integrated sensing systems that can combine several types of data in one module, reducing the need for robot makers to buy and connect separate sensors.
“In the robotics industry, multi-sensing — the integration of numerous sensors into a single system — is expected to become increasingly important,” LG Innotek CEO Moon Hyuksoo noted. “Through our collaboration with TDK, which offers a broad portfolio of sensor technologies, we aim to help shape the ecosystem for core robotic components.”
The first planned product is a vision sensing module and LG Innotek said it expects to launch a module with an inertial sensor in 2027.
Vision Module
The planned next-generation vision module is designed to improve robot perception by combining image data with information about movement, position and sound. LG Innotek said that approach would give robots more context than image sensors alone.
One planned version will integrate TDK’s inertial measurement unit, or IMU, into the vision module. The companies expect that setup to help compensate for motion blur, stabilize images and reduce blind spots through high-frequency tracking and dead reckoning.
Tactile Sensing
The companies also plan to develop a tactile sensing module by 2027. The module would function like robot skin, allowing robots to detect contact and pressure when their hands, arms or bodies touch objects. LG Innotek said it will contribute module design and digital signal-processing capabilities, while TDK will provide sensor technology from its consumer, automotive and industrial work.
The companies also plan to expand cooperation into real-time processing and sensor-data analysis.
LG Innotek is already working in physical AI sensing through composite modules that combine camera modules, LiDAR and radar. The company said it is also working with other technology companies, including Boston Dynamics. In June, Nvidia announced LG will use Nvidia technologies to develop humanoid, logistics and industrial robots and build a physical AI data factory.
Image: John Min, LG Innotek CTO (right), and Shuichi Hashiyama, TDK CTO, pose for a commemorative photo at the signing ceremony of a Memorandum of Understanding for cooperation in developing physical AI solutions, held at LG Innotek’s headquarters in Magok, Gangseo-gu, Seoul. (Credit: LG Innotek)
