Tech news in 3 minutes
Robotaxis drive miles just to get cleaned and charged; this new startup wants to fix that
San Francisco-based startup Aseon Labs has raised $10 million in a seed round led by Crane Venture Partners to develop automated pods that inspect, clean, and charge autonomous vehicles, aiming to reduce "deadhead miles" (miles driven without passengers) and improve robotaxi profitability. The round also saw participation from Y Combinator, Uber co-founder Garrett Camp’s Expa, Robin Hood Ventures, Founders Capital, and angel investors including former Google executive Adrian Aoun, Mercury CEO Immad Akhund, Zimride co-founder Rajat Suri, and team members from Anthropic, Nuro, Turo, and Revolut. The Redwood City-based company was founded by George Kalligeros (CEO) and Dan Keene (COO), who previously built battery-swapping startup Pushme, acquired by Tier Mobility in 2020. They apply the same "non-permanent infrastructure" approach to robotaxi depots. Currently, robotaxi companies rely on central depots outside city centers for maintenance and charging, leading to high real estate costs and empty miles. Aseon’s solution: parking space-sized autonomous pods that can be scattered across cities. These temporary structures include cameras and robotic arms to inspect vehicles, clean interiors, and retrieve lost items. They run on propane generators or can connect to existing power. While early versions will be staffed, the goal is full autonomy. Computer vision and AI (vision-language-action models) help the system avoid tackling problems it can't solve—like melted chocolate stains—by dispatching the vehicle to a human-staffed depot instead. The seed funds will build five prototype pods, grow the six-person team to about a dozen, and secure real estate for the network. Aseon has no signed contracts yet but says robotaxi operators are interested in trying the concept. Kalligeros stressed that reducing deadhead miles and increasing vehicle utilization is key to achieving economic parity with traditional ride-hailing.