MIT engineers have developed a new training method to help ensure the safe operation of multiagent systems, including robots, search-and-rescue drones and self-driving cars, reports Jijo Malayil for Interesting Engineering. The new approach “doesn’t focus on rigid paths but rather enables agents to continuously map their safety margins—the boundaries within which they must stay,” writes Malayil.
Writing for Bloomberg, David Zipper, senior fellow at the MIT Mobility Initiative, highlights the impact of the robotaxi industry on public transportation. “Transit-robotaxi synergy is an enticing message at a time when public transportation agencies face a dire funding shortage, and it could especially resonate among left-leaning residents in places like the Bay Area who value buses and trains even if they seldom use them,” writes Zipper. “But caveat emptor: The robotaxi industry’s embrace of public transportation conceals a wolf in sheep’s clothing.”
Researchers at MIT have developed “Clio,” a new technique that “enables robots to make intuitive, task-relevant decisions,” reports Jennifer Kite-Powell for Forbes. The team’s new approach allows “a robot to quickly map a scene and identify the items they need to complete a given set of tasks,” writes Kite-Powell.
Researchers at MIT have developed a new method that “enables robots to intuitively identify relevant areas of a scene based on specific tasks,” reports Baba Tamim for Interesting Engineering. “The tech adopts a distinctive strategy to make robots effective and efficient at sorting a cluttered environment, such as finding a specific brand of mustard on a messy kitchen counter,” explains Tamim.
A new study by researchers from MIT and elsewhere has found that “most drivers are likely to multitask and get distracted if their vehicles are equipped with partial automation software,” reports William Gavin for Quartz. The researchers “studied how drivers with cars that have the technology behaved after it was enabled,” explains Gavin.
Edwin Olson '00, MEng '01, PhD '08 founded May Mobility, an autonomous vehicle company that uses human autonomous vehicle operators on its rides, reports Gus Alexiou for Forbes. “May Mobility is focused above all else on gradually building up the confidence of its riders and community stakeholders in the technology over the long-term,” explains Alexiou. “This may be especially true for certain more vulnerable sections of society such as the disability community where the need for more personalized and affordable forms of transportation is arguably greatest but so too is the requirement for robust safety and accessibility protocols.”
Axios reporter Alex Fitzpatrick spotlights MightyFly, an aviation startup founded by Manal Habib ’11 that is developing a large, autonomous electric vehicle takeoff and landing cargo drone that has been approved by the Federal Aviation Administration for a flight corridor. "The use case is B2B expedited logistics," says Habib. "Think of deliveries from a manufacturer to suppliers. Think of deliveries from a lab to a hospital, or from a warehouse or pharmacy, as well as to improve deliveries to an oil rig or to a farm or a mining site, as well as for DOD use cases."
David Zipper, Senior Fellow at the MIT Mobility Initiative, speaks with Washington Post reporter Trisha Thadani about the safety behind self-driving car companies, such as Google’s robotaxi service, Waymo. Zipper says there is a disparity that “the companies are saying the technology is supposed to be a godsend for urban life, and it’s pretty striking that the leaders of these urban areas don’t really want them.”
Research scientist Bryan Reimer speaks with The Ojo-Yoshida Report host Junko Yoshida about the future of the autonomous vehicle industry. “We cannot let the finances drive here,” explains Reimer. “We need to manage the finances to let society win over the long haul.”
Forbes reporter Rob Toews spotlights Prof. Daniela Rus, director of CSAIL, and research affiliate Ramin Hasani and their work with liquid neural networks. “The ‘liquid’ in the name refers to the fact that the model’s weights are probabilistic rather than constant, allowing them to vary fluidly depending on the inputs the model is exposed to,” writes Toews.
Venti Technologies, which was co-founded by MIT researchers and alumni, is working to build autonomous vehicles for industrial and global supply chain hubs, reports Bruce Rogers for Forbes. “Working with the world's leading port operator provides Venti the opportunity to bring the economics of autonomous vehicles to over 60 ports globally,” writes Rogers. “These ports operate 24/7 requiring 2-3 shifts of human drivers.”
Researchers at MIT have developed a new artificial intelligence system aimed at helping autopilot avoid obstacles while maintaining a desirable flight path, reports Kyle Wiggers for TechCrunch. “Any old algorithm can propose wild changes to direction in order to not crash, but doing so while maintaining stability and not pulping anything inside is harder,” writes Wiggers.
Popular Science reporter Jamie Dickman writes that using liquid neural networks, MIT researchers have “trained a drone to identify and navigate toward objects in varying environments.” Dickman notes that: “These robust networks enable the drone to adapt in real-time, even after initial training, allowing it to identify a target object despite changes in their environment.”
Researchers at MIT have developed a new type of autonomous drone that uses advanced neural networks to fly, reports Tony Ho Tran for The Daily Beast. “The new design allows the drone to make better decisions when flying through completely new environments,” writes Tran, “and could have future applications in self-driving cars, search and rescue operations, wildlife monitoring, or even diagnosing medical issues.”
MIT researchers have developed Robust MADER, an updated version of a previous system developed in 2020 to help drones avoid in-air collisions, reports Brian Heater for TechCrunch. “The new version adds in a delay before setting out on a new trajectory,” explains Heater. “That added time will allow it to receive and process information from fellow drones and adjust as needed.”