The coordination of multiple autonomous agents in high-speed, competitive environments represents a significant engineering challenge. This paper presents CRUISE (Curriculum-Based Iterative Self-Play for Scalable Multi-Drone Racing), a reinforcement learning framework designed to solve this challenge in the demanding domain of multi-drone racing. CRUISE overcomes key scalability limitations by synergistically combining a progressive difficulty curriculum with an efficient self-play mechanism to foster robust competitive behaviors. Validated in high-fidelity simulation with realistic quadrotor dynamics, the resulting policies significantly outperform both a standard reinforcement learning baseline and a state-of-the-art game-theoretic planner. CRUISE achieves nearly double the planner's mean racing speed, maintains high success rates, and demonstrates robust scalability as agent density increases. Ablation studies confirm that the curriculum structure is the critical component for this performance leap. By providing a scalable and effective training methodology, CRUISE advances the development of autonomous systems for dynamic, competitive tasks and serves as a blueprint for future real-world deployment.

Urban Air Mobility (UAM) envisions the widespread use of small aerial vehicles to transform transportation in dense urban environments. However, UAM faces critical operational challenges, particularly the balance between minimizing noise exposure and maintaining safe separation in low-altitude urban airspace, two objectives that are often addressed separately. We propose a reinforcement learning (RL)-based air traffic management system that integrates both noise and safety considerations within a unified, decentralized framework. Under this scalable air traffic coordination solution, agents operate in a structured, multi-layered airspace and learn altitude adjustment policies to jointly manage noise impact and separation constraints. The system demonstrates strong performance across both objectives and reveals tradeoffs among separation, noise exposure, and energy efficiency under high traffic density. The findings highlight the potential of RL and multi-objective coordination strategies in enhancing the safety, quietness, and efficiency of UAM operations.

Cruise robotaxis are back on the road… well, kind of. Though General Motors pulled the plug on its self-driving taxi business last year, the automaker has been quietly repurposing a few of the vehicles as it seeks to develop new driver-assistance technologies. This week, WIRED spotted a GM Bolt electric hatchback on the San Francisco-Oakland Bay Bridge, and later saw a similar vehicle on Interstate 880 near Oakland. In each instance, the car was being driven by a human. The vehicle had "Mint" written on the hood, but didn't include any visually apparent Cruise branding.

Automakers and tech developers testing and deploying self-driving and advanced driver assistance features will no longer have to report as much detailed, public crash information to the federal government, according to a new framework released today by the US Department of Transportation. The moves are a boon for makers of self-driving cars and the wider vehicle technology industry, which has complained that federal crash reporting requirements are overly burdensome and redundant. But the new rules will limit the information available to those who watchdog and study autonomous vehicles and driver assistance features--tech developments that are deeply entwined with public safety but which companies often shield from public view because they involve proprietary systems that companies spend billions to develop. The government's new orders limit "one of the only sources of publicly available data that we have on incidents involving Level 2 systems," says Sam Abuelsamid, who writes about the self-driving vehicle industry and is the vice president of marketing at Telemetry, a Michigan research firm, referring to driver assistance features such as Tesla's Full Self-Driving (Supervised), General Motors' Super Cruise, and Ford's Blue Cruise. These incidents, he notes, are only becoming "more common."

Accurately quantifying air-sea fluxes is important for understanding air-sea interactions and improving coupled weather and climate systems. This study introduces a probabilistic framework to represent the highly variable nature of air-sea fluxes, which is missing in deterministic bulk algorithms. Assuming Gaussian distributions conditioned on the input variables, we use artificial neural networks and eddy-covariance measurement data to estimate the mean and variance by minimizing negative log-likelihood loss. The trained neural networks provide alternative mean flux estimates to existing bulk algorithms, and quantify the uncertainty around the mean estimates. Stochastic parameterization of air-sea turbulent fluxes can be constructed by sampling from the predicted distributions. Tests in a single-column forced upper-ocean model suggest that changes in flux algorithms influence sea surface temperature and mixed layer depth seasonally. The ensemble spread in stochastic runs is most pronounced during spring restratification.

Autonomous vehicle company Cruise is laying off around half of its workforce, according to reporting by TechCrunch. The cuts even extend to the CEO and other top executives. This is part of a major restructuring effort by parent company GM that will eventually see a total shutdown of operations. These layoffs are expected to impact well over 1,000 people and include CEO Marc Whitten, chief safety officer Steve Kenner and global head of public policy Rob Grant. Chief technologist Mo Elshenawy is also being laid off, but will stay on until the end of April to help with the transition.

It looks like a cross between a toy airplane and a drone, but this British solar-powered aircraft could be the future of aerial surveillance. PHASA-35, built by British company BAE Systems, is a 150kg solar-electric aircraft that can quietly cruise through the stratosphere for months at a time. Named after its 35-metre wingspan, the unmanned aerial vehicle (UAV) travels at a maximum height of 70,000 feet, at a leisurely speed of 55mph. Designed as a cheaper and lighter alternative to satellites, it can be used for Earth observation and surveillance, border control, communications and disaster relief. Now, BAE Systems reveals that PHASA-35 has just completed a second round of test flights into the stratosphere – the second layer of Earth's atmosphere.

This week in tech: General Motors says goodbye to robotaxis but not self-driving cars; one woman's fight to keep AI out of applications for housing; Salt Typhoon; and tech's donations to Donald Trump. Thank you for joining me. When God shuts down one robotaxi business, he resurrects another. Last week, General Motors announced it would stop funding its subsidiary Cruise, which made self-driving car software and operated a robotaxi service. The unit had been a leader in autonomous vehicles until a near-fatal crash in late 2023, when a Cruise car hit a pedestrian and dragged her along the road underneath its chassis.

General Motors is officially ending its support for Crusie's beleaguered fleet of self-driving "robotaxis." In a surprise announcement this week, the US carmaker said it will "realign its autonomous driving strategy" to end robotaxis and instead focus on eventually creating an autonomous personal vehicle. Cruise, which previously operated as a subsidiary, will now be fully absorbed by GM. That's all a major departure for the driverless car company which had its sights set on offering paid robotaxis rides in multiple cities next year. Cruise previously proclaimed it planned to have close to a million of its autonomous vehicles flooding US streets by the end of the decade.

GM attributed the change of strategy to "the considerable time and resources that would be needed to scale the business". The company did not say how many Cruise employees could be moved over to GM. GM, which owns about 90% of Cruise, said it has agreements with other shareholders that will raise its ownership to more than 97%. In December 2023, Cruise said it would cut 900 jobs, about a quarter of its workforce. The announcement came as safety officials were investigating the firm after reports of injuries to pedestrians. Cruise had earlier pulled all of its US vehicles from testing after California halted its driverless testing permit.