An Ultrasound ASIC With Universal Energy Recycling for >7-m All-Weather Metamorphic Robotic Vision

Abstract

An ultrasound ASIC that enables compact, 3-D nonvisual robotic navigation for >7-m is presented. The proposed ultrasound ASIC mitigates the limitations of camera vision navigation systems in all-weather, low-power, low-cost aspects. The ASIC integrates TX and RX paths for 64 channels, enabling four-directional navigation using 4 × 4 ultrasound arrays. The proposed universal energy recycling transmitter (UERTX) driver circuit, which can drive both single-ended and differential transducers, reduces energy consumption by 44% compared with non-overlap switching-assisted conventional class-D ultrasonic drivers, addressing large TX power consumption. In addition, the on-chip programmable power management unit (PMU) eliminates off-chip supplies and reduces off-chip passive components for compact system miniaturization. Furthermore, PMU enables the tuning of TX driving amplitude for the range of 6-14 V pp allowing wide detection range adaptability. Fabricated in 0.18 µm 1P6M standard CMOS, the ASIC system measures 25 mm 2 and has >7-m obstacle detection capability while consuming 4.3 mW/channel. The proposed ultrasound ASIC system is demonstrated with a four fps, 3-D navigating metamorphic robot, which consumes 0.28 W, occupies 125 cm 3 , and weighs ≤100 g.

An Ultrasound Imaging System With On-Chip Per-Voxel RX Beamfocusing for Real-Time Drone Applications

“…Ultrasound images are traditionally rendered by beamforming the transmitting or the receiving acoustic waves, transforming the echo strength envelope, and scan-converting the signal to images. In the previous works [15], [16], [17], [18], only prebeamforming steps are integrated into the analog domain, and the remaining image reconstruction steps, parallel postbeamforming till scan-conversion, are processed on off-chip components. The off-chip components constrain the overall system power and form factor.…”

“…Although the charge-recycling high-voltage TX (CRHV-TX) designs proposed in [10] and universal energy recycling TX [16] improved the output swing to 13.2 and 14 V pp , respectively, they are still insufficient for the 7-m-range object detection. As shown in Fig.…”

“…Charge pump-based boost conversion [16] (with open-loop control) requires a fixed switching frequency for the peak power drain and increasing hard-charging loss and switching overhead. A flying capacitor cross-connected dc-dc boost converter proposed in [28] can reduce the V DS stress of some transistors with a high conversion ratio and single high-voltage output.…”

“…Moreover, a LiDAR system may require a power-hungry transmitter (TX), such as a >40-W pulsed laser diode [6], [7]; and it is also an expensive solution, making it less attractive for consumer-grade drone applications. Ultrasound imaging [10], [11], [12], [13], [14], [15], [16], [17], [18], on the other hand, can provide low-power 3-D-depth information and works well even in complex lighting conditions (bright/dark) as well as in rainy/foggy weather. Of course, this comes at the cost of lower spatial resolution when compared to CIS or LiDAR, but this is not a major issue in drone navigation.…”

“…Previously, UISs [10], [11], [12], [13], [14] shown successful results for biomedical applications. Due to the slow sound speed (343 m/s in air compared to 1540 m/s in tissue) and long detection range compared to in vivo, air-channel ultrasound imaging [15], [16], [17], [18] possesses a different challenge in how to achieve adequate framerate with a limited number of transmissions. Prior study [19] discussed techniques that could 1 m and it does not integrate on-chip image reconstruction 87 DBE; another work [18] demonstrated real-time ultrasound 88 imaging in air channel up to 3 m range at 29 frames/s, 89 but the system is made up of discrete components (no chip 90 implementation), making its bulky form factor and high power 91 consumption (10 W) beyond what a common UAV can 92 handle.…”