TDK InvenSense Enables Smart Canes, Indoor Positioning

Published: January 17, 2025
  • The sensor technology company released four product announcements at CES 2025 that demonstrated how the company’s latest products can improve situational awareness
  • In addition to WeWALK smart canes and 9-Axis positioning for indoor navigation known as PositionSense, its new products include VibeSense360 to enhance use of audio, and SmartSonic ultrasonic time-of-flight for smart homes

TDK InvenSense announced four new sensor technology innovations for consumer devices and for developers to bring new features to IoT devices. The new solutions offer more precise sensor data and lower power consumption, according to company officials.

It is offerings its 9-Axis PositionSense to help ensure users of consumer devices are never lost, even indoors. Its next-generation VibeSense360 provides more immersive sound. The company’s ultrasonic SmartSonic time-of-flight (ToF) sensor helps enhance range and accuracy for smart home applications, while TDK also has collaborated with WeWALK for the UK company’s Smart Cane 2, to track conditions and improve sensory based awareness, with machine learning, for the visually impaired.

The company demonstrated its new technology at this year CES show.

Smart Cane Tracks Conditions

The WeWALK Smart Cane 2 leverages TDK’s micro-electro-mechanical systems (MEMS) microphones, inertial measurement units (IMUs) and ultrasonic time-of-flight sensors for voice prompts, navigation and obstacle avoidance, explained Massimo Mascotto, TDK’s product marketing director.

Founded in 2019, WeWALK is a UK-based startup that uses technology to improve mobility for visually impaired people. It leverages sensor-based technology with AI, built into a white cane used by those walking with a visual impairment.

The new cane uses TDK’s voice assistance so that users can press a button and ask a question. TDK’s sensors provide overhead obstacle detection as well, said Mascotto. The sensor helps identify when a person or obstacle is situated in front of the individual with the cane, in a position where the cane does not touch.

SmartSonic for Drones, Thermostats and Robotics

To accomplish the obstacle detection, the TDK InvenSense SmartSonic ICU-30201 ultrasonic ToF sensor transits and detects responses from objects around it. Mascotto detailed that the ICU-30201 employs ultrasonic as opposed to optical sensors for presence detection, without the higher battery demands that come with optical technology.

Beyond the Smart Cane 2, the technology is being used in smart devices to bring intelligence about occupancy, movement and distance measurements within range of the device.

Applications include detection of individuals or activities for smart building applications, said Song Li, director of product marketing at TDK InvenSense. The sensor determines the presence of a person and a response can be triggered, such as turning up the heat or air conditioning. Another application is obstacle avoidance for indoor robotics and drones.

PositionSense with 9 Axis

For indoor or precise navigation that expands on standard GPS, the company’s PositionSense is designed for use with wearables, drones and robotics. It comes with a 6-axis SmartMotion Inertial measurement unit (IMU) including a 3-axis gyroscope and 3-axis accelerometer.

Additionally, it includes tunnel-magneto resistance (TMR) and Pedestrian Dead Reckoning (PDR).

Users with the technology in their smartphone or wearable device could use the system to understand their location and position in an indoor and crowded location, such as a subway station. As the train arrives in the station, doors open, and the user steps out of the train and onto the platform, they can see their position, where the exit or stairs are, or another platform and follow directions via an app.

PositionSense comes with “always on” gyro assist magnetometer calibration for real time location without high battery consumption.

Intelligent Audio

Lastly, TDK’s VibeSense360 aids with true wireless stereo (TWS) for individuals wearing earbuds or headphones. The technology provides spatial head tracking for more immersive and ultra-low-power spatial audio experience depending on the environment.

Those features include Transparency For Talk “to detect when the user is speaking through bone conduction,” said Li. The system can automatically shift from noise cancellation to transparency mode, as well as pause any audio that may be playing. In other words, if the individual wearing the earbuds or headphone starts speaking, music can go quiet or noise cancellation from the exterior environment can be halted.

The Keyword Assist feature assists with always-on microphones in detecting human voice activity through bone conduction, ensuring more secure, accurate and ultra-low-power keyword detection while avoiding false triggers—even in noisy environments. Users don’t need to tap their earbuds, but simply tap their cheek, for instance.

And UI Gestures or Activity Detection enables a smoother user experience through always-on gesture detections like tap, double tap, triple tap, wide-area tap, head gestures like shake and nod, and activity or inactivity detection. That feature can trigger fitness and health monitoring.

Machine Learning for Sensors

TDK’s machine learning algorithms are built to run inside the IMU, to recognize customized activities and gestures, assigning commands or actions based on a user’s own motions.

InvenSense sensors are built to reduce the overall system power consumption and that—in some applications—can also save on the latency related to transmitting a large amount of data. That means managing some data on the edge, Mascotto explained.

And when it comes to machine learning, developers or users can access InvenSense’s machine learning software from its website, choose certain filters and then they can train it based on the activities it will be exposed to.

Last month the sensors were being produced in large volume, such as robotics, drones, alarm devices, door locks, camera devices and smart building systems.

Learn More: