Obstacle avoidance

Thanks to their highly sensitive whiskers, which enable them to detect objects within reach before they get too close, rats and mice can hardly relate to the pain of a stubbed toe, despite having eight more toes than humans. This sensory organ lets rats scurry through buildings, sewers, burrows, and countless other environments with ease, regardless of visibility. Beyond touch, whiskers can also detect vibrations and even subtle air currents.

While in-vehicle collision avoidance systems may be good at detecting big targets like automobiles, they're not as adept at detecting bicycles. Radian's Wingman was designed to address that problem, by boosting bikes' radar and LiDAR visibility.

A Singapore startup recently launched an e-mobility concept that looks like an unhappy marriage between an office chair and a mobility scooter. The Strutt ev1 sports a sensor array and smart algorithms that feed into a built-in drive-assist system.

The first 5G-enabled ebike has rolled into Mobile World Congress in Barcelona this week, courtesy of Long Island-based mobile device company Orbic. The fat-tire pedal-assist ride also comes with a bunch of cameras and AI crash detection.

Self-driving cars already utilize cameras to avoid collisions with other vehicles or pedestrians. Well, the biped wearable device applies that same principle to guiding the blind as they walk along city streets.

Samsung is claiming a few world firsts with its latest robot vacuum cleaner. The Jet Bot AI+ is said to be the first to feature an active stereo-type 3D sensor for object detection, and the first equipped with Intel AI for object recognition.

Manufactured by Austrian startup Tec-Innovation, the InnoMake shoe uses ultrasound sensors to warn blind users of obstacles in their path. The footwear may soon become even more capable, though, thanks to integrated cameras.

While autonomous drones are now pretty good at avoiding large obstacles, thin suspended power lines can still pose a challenge. A new onboard sensing system, however, could change that.

Obstacle avoidance is crucial for drones, but most systems aren’t fast enough for some situations. Now engineers at the University of Zurich have developed a new system that gives drones such fast reflexes that they can play – and win at – dodgeball.

Our experiences with so-called self-flying drones so far have fallen a little short of the promised entirely handsfree form of fantastical film-making, but if the demo footage that accompanies the newly announced Skydio R1to go by, that could soon change. ​​

Parrot's newly-announced S.L.A.M.dunk is a development kit for drones equipped with hardware that suggests a particular focus on obstacle avoidance and navigation, but doubles as a desktop computer as well. ​

Among the many computer scientists working on crash avoidance systems for drones is a team of MIT researchers, who have developed route-planning software that allows the aircraft to make intricate turns to autonomously navigate tight spaces. ​​