SoundVision - An ultrasonic 3D distance sensor for autonomous robots

In SoundVision, Sonair AS will develop a new product, a 3D distance sensor type that provides autonomous robots with omnidirectional depth perception to understand its surroundings. Mobile robots constitutes a huge and growing market. For robots that interact with humans, it is critical and enabling that the robot has reliable control of its surroundings in all possible directions of movement. Our sensor will make it possible to develop cost-effective autonomous robots that can safely work alongside humans or be positioned in close proximity to humans. We call this new sensor type ADAR™ (Acoustic Detection And Ranging). ADAR™ has only recently become possible to realize using MEMS (Micro Electro-Mechanical System)-based ultrasound transducers. In air, commercial ultrasound transducers are too large to obtain the fundamental half-wavelength element spacing needed. SINTEF has over the last decade worked to scale down ultrasound transducers by using a MEMS technology called PMUT (Piezoelectric Micromachined Ultrasonic Transducer), that show comparable performance to commercial bulk devices at 1/10 the size. The imaging method is called software beamforming. Beamforming is the backbone processing of SONAR and RADAR (RAdio Detection And Ranging), as well as in medical ultrasound imaging. Beamforming is to steer the signals received and/or transmitted on/from an array into a certain direction or location. In this way, the sensor system will provide 3D information in real-time. Recent hardware developments have allowed the implementation of beamforming on received signals as a post-processing method; software beamforming. The innovation is to combine wavelength matched PMUTs, cutting-edge software beamforming and object recognition algorithms, because 3D spatial information is available by simply transmitting sound and listening.

In SoundVision, Sonair AS will develop a new product, a 3D distance sensor type that provides autonomous robots with omnidirectional depth perception to understand its surroundings. Mobile robots constitutes a huge and growing market. For robots that interact with humans, it is critical and enabling that the robot has reliable control of its surroundings in all possible directions of movement. Our sensor will make it possible to develop cost-effective autonomous robots that can safely work alongside humans or be positioned in close proximity to humans. We call this new sensor type ADAR™ (Acoustic Detection And Ranging). ADAR™ has only recently become possible to realize using MEMS (Micro Electro-Mechanical System)-based ultrasound transducers. In air, commercial ultrasound transducers are too large to obtain the fundamental half-wavelength element spacing needed. SINTEF has over the last decade worked to scale down ultrasound transducers by using a MEMS technology called PMUT (Piezoelectric Micromachined Ultrasonic Transducer), which show comparable performance to commercial bulk devices at 1/10 the size. The imaging method is called software beamforming. Beamforming is the backbone processing of SONAR and RADAR (RAdio Detection And Ranging), as well as in medical ultrasound imaging. Beamforming is to steer the signals received and/or transmitted on/from an array into a certain direction or location. In this way, the sensor system will provide 3D information in real-time. Recent hardware developments have allowed the implementation of beamforming on received signals as a post-processing method; software beamforming. The innovation is to combine wavelength matched PMUTs, cutting-edge software beamforming and object recognition algorithms, because 3D spatial information is available by simply transmitting sound and listening.