Simulation-based testing represents an important step to ensure the reliability of autonomous driving software. In practice, when companies rely on third-party general-purpose simulators, either for in-house or outsourced testing, the generalizability of testing results to real autonomous vehicles is at stake. In this paper, we strengthen simulation-based testing by introducing the notion of digital siblings, a novel framework in which the AV is tested on multiple general-purpose simulators, built with different technologies. First, test cases are automatically generated for each individual simulator. Then, tests are migrated between simulators, using feature maps to characterize of the exercised driving conditions. Finally, the joint predicted failure probability is computed and a failure is reported only in cases of agreement among the siblings. We implemented our framework using two open-source simulators and we empirically compared it against a digital twin of a physical scaled autonomous vehicle on a large set of test cases. Our study shows that the ensemble failure predictor by the digital siblings is superior to each individual simulator at predicting the failures of the digital twin. We discuss several ways in which our framework can help researchers interested in automated testing of autonomous driving software.

This paper proposes a reinterpretation of selective attention as a form of control of working memory based on self-generated reward signals and model-free reinforcement learning.

Ordering your lunch or coffee using a self-serve computer screen instead of speaking to a human is one of the most obvious examples of how the workplace and retail experience is becoming more automated. Such automation has become so common that Starbucks is taking steps to make sure the process doesn't feel so, well, robotic. Starbucks announced this month that it was installing two-way video screens at its drive-thus to personalize the transaction, "allowing customers and baristas to see each other and truly interact when the order is being placed." Starbucks also said its mobile order and pay program, which allows customers to order through a phone app and swoop in to pick up an order without waiting in line, also provides "personalized customer experiences." Other fast food outlets, including A&W and McDonald's, also give customers the option to bypass or minimize human interaction with self-serve screen kiosks, and experts predict the automation of work tasks is likely going to speed up in coming years.

It looks like Spock and Darth Vader may have some competition in the mind control stakes. Researchers in the U.S. claim to have discovered the secret to mind control by creating the world's first ever human-to-human brain interface. Using electrical brain recordings and a form of magnetic stimulation, Rajesh Rao sent a brain signal to Andrea Stocco on the other side of the Washington University campus, causing Stocco's finger to move on a keyboard. University of Washington researcher Rajesh Rao, left, plays a computer game with his mind. Across campus, researcher Andrea Stocco, right, wears a magnetic stimulation coil over the left motor cortex region of his brain.