Up, up and away: Dubai hopes to have a passenger-carrying drone regularly buzzing through the skyline of this futuristic city-state in July. The arrival of the Chinese-made EHang 184 -- which already has had its flying debut over Dubai's iconic, sail-shaped Burj al-Arab skyscraper hotel -- comes as the Emirati city also has partnered with other cutting-edge technology companies, including Hyperloop One. The question is whether the egg-shaped, four-legged craft will really take off as a transportation alternative in this car-clogged city already home to the world's longest driverless metro line. Mattar al-Tayer, the head of Dubai's Roads & Transportation Agency, announced plans to have the craft regularly flying at the World Government Summit. Before his remarks on Monday, most treated the four-legged, eight-propeller craft as just another curiosity at an event that views itself as a desert Davos.
As I approached San Francisco International Airport, my expectations for BMW's new concept car were as big as the looming Boeing 777F Lufthansa cargo jet waiting for me. I had surrendered my cellphone and everything in my purse but my drivers license to see BMW's iNext vehicle. Its tour started in Munich a few days earlier; it came to the Bay Area after a stop at New York's JFK airport, and was scheduled to continue on to Beijing. SEE ALSO: BMW makes sure we can't escape voice assistants while driving After passing a final security check, I climbed up the rickety staircase with fellow media members and entered the cavernous aircraft. We had been told very little about what we were going to see, except it was not only the "car of the future" but the "idea of the future."
French inventor Frank Zapata grabbed headlines around the world this summer when he flew his hoverboard across the English channel from Pas de Calais, France, to the famous white cliffs of Dover. But Bay Area commuters may soon do Zapata one better by skimming above San Francisco Bay on autonomous, single-passenger drones being developed by a Peninsula start-up company with ties to Google. The automated drones are electrically powered, capable of vertical takeoff and landing, and would fly 10 feet above the water at 20 mph along a pre-determined flight path not subject to passenger controls. The drones' rotors are able to shift from vertical to horizontal alignment for efficient forward movement after takeoff. The company behind all this, three-year-old Kitty Hawk Corp., has personal financial backing from Google founder Larry Page, now CEO of Google's parent, Alphabet, who has long been interested in autonomous forms of transportation.
With its Shinkansen'bullet trains' and melodious subway system, Tokyo already has some of the world's greatest public transport infrastructure. But the heavily populated city will be pushed to its limits come 2020, when the world descends on the Japanese capital as it plays host to the Olympic games. One company looking to capitalise on the influx of tourists is robotics firm ZMP Inc. According to Reuters, it's planing to team up with Tokyo's Hinomaru Kotso cab firm to update its fleet of 600 cars with driverless technology. ZMP has already had driverless cars on Tokyo's streets, but each had a driver ready to wrestle control should the AI go wayward.
Object detection models shipped with camera-equipped mobile devices cannot cover the objects of interest for every user. Therefore, the incremental learning capability is a critical feature for a robust and personalized mobile object detection system that many applications would rely on. In this paper, we present an efficient yet practical system, IMOD, to incrementally train an existing object detection model such that it can detect new object classes without losing its capability to detect old classes. The key component of IMOD is a novel incremental learning algorithm that trains end-to-end for one-stage object detection deep models only using training data of new object classes. Specifically, to avoid catastrophic forgetting, the algorithm distills three types of knowledge from the old model to mimic the old model's behavior on object classification, bounding box regression and feature extraction. In addition, since the training data for the new classes may not be available, a real-time dataset construction pipeline is designed to collect training images on-the-fly and automatically label the images with both category and bounding box annotations. We have implemented IMOD under both mobile-cloud and mobile-only setups. Experiment results show that the proposed system can learn to detect a new object class in just a few minutes, including both dataset construction and model training. In comparison, traditional fine-tuning based method may take a few hours for training, and in most cases would also need a tedious and costly manual dataset labeling step.