惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Y
Y Combinator Blog
GbyAI
GbyAI
爱范儿
爱范儿
H
Hackread – Cybersecurity News, Data Breaches, AI and More
C
Check Point Blog
M
MIT News - Artificial intelligence
量子位
宝玉的分享
宝玉的分享
MongoDB | Blog
MongoDB | Blog
V
Visual Studio Blog
罗磊的独立博客
F
Fortinet All Blogs
美团技术团队
博客园_首页
博客园 - 【当耐特】
L
LangChain Blog
月光博客
月光博客
腾讯CDC
The Cloudflare Blog
D
Docker
博客园 - 聂微东
Stack Overflow Blog
Stack Overflow Blog
WordPress大学
WordPress大学
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报

TechSpot

Flagship Rematch: Ryzen 7 5800X3D vs. Core i9-12900K Slack chats and internal data from failed startups are finding a second life in AI training A $5 Bluetooth tracker hidden in a postcard exposed a warship's movements Leakers claim PlayStation 6 could offer at least 3x the performance of the PS5 The Mac Mini is no longer a niche product, it's local AI infrastructure IPv6 traffic reaches parity with IPv4 for the first time, Google data shows Xbox expansion cards are now cheaper than SSDs, and PC users are repurposing them Blue Origin prepares to reuse New Glenn booster in bid to challenge SpaceX Nvidia could bring back the 12GB RTX 3060 as supply issues disrupt GPU roadmap What was the first OS you ever used? SNK revives NeoGeo AES with modern upgrades and HDMI support Valve's Proton 11 beta boosts Linux gaming with better performance and classic game support Researchers warn Microsoft Defender vulnerability is already being exploited A four-day Steam freebie turned into $250,000 for an indie game AMD may relaunch Ryzen 7 5800X3D for AM4's 10th anniversary This humanoid robot can almost run as fast as a human sprinter Two New Jersey men jailed for helping North Korean IT workers infiltrate 100+ companies A $7,000 DIY radar project is taking on hardware that usually costs over $100,000 Metro 2039 is going darker than ever, launching this winter on PC and consoles Gemini arrives on macOS with a dedicated desktop app AI infrastructure boom pushes AMD, Intel and Arm to new valuation heights New self-healing material can repair itself over 1,000 times, extend the lifespan of cars and aircraft Japan's bullet train to debut high-tech private cabins, for an added fee Memory card and flash drive pricing surges 120%, with some models spiking 260% Open-source tool decrypts all private data collected by Windows Recall on Copilot PCs The 2026 PC and Console Gaming Report shows most revenue now comes from games outside the Top 20 PureMac is a new open-source macOS cleanup and app removal tool Your Airbnb host might actually be AI Steam might soon display 30-day price history for game deals Intel brings 18A process to budget laptops with new Core Series 3 CPUs
Waymo recalls 3,871 Robotaxis after cars drove into activ...
Skye Jacobs · 2026-06-19 · via TechSpot

Serving tech enthusiasts for over 25 years.
TechSpot means tech analysis and advice you can trust.

What just happened? The challenge for autonomous vehicles has never been ordinary driving – it's the edge cases. Temporary roadwork, shifting lane lines and irregular signage can still confuse even sophisticated systems. A recent recall by Waymo shows how those gaps can surface in real-world deployments.

Waymo is recalling nearly 3,900 robotaxis following incidents in freeway construction areas. According to a safety recall report filed with the National Highway Traffic Safety Administration, the company's fifth-generation automated driving system may enter and drive at speed in freeway construction zones under certain circumstances, either because it fails to recognize the construction zone or prioritizes other freeway hazards.

The filing identifies 3,871 vehicles as potentially affected. It states that under certain circumstances, the system "may enter and drive at speed in freeway construction zones due to inappropriately prioritizing the avoidance of other freeway hazards and/or failing to recognize the construction zone."

The problem was not confined to a single event. On April 11, a Waymo vehicle was involved in an incident on a freeway, prompting the company to review how the system handled closures. Five similar events followed on April 19 in Phoenix. On May 18, seven vehicles in the San Francisco Bay Area drove between cones and into freeway lanes where construction was underway, a pattern that suggests difficulty reading temporary lane layouts.

Together, the incidents suggest a broader issue with how the system processes dynamic environments rather than a one-off failure. Construction zones are inherently inconsistent. Signage can vary, cones can shift, and lane configurations often change. For human drivers, those conditions demand judgment calls. Automated systems require accurate detection paired with real-time decision-making that can adapt.

Waymo responded by restricting freeway driving in affected scenarios while it worked on a fix. The company's field safety committee put those limits in place shortly after the April incidents, and its safety board formally approved the recall on June 8 after reviewing additional data.

The remedy will focus on software updates. According to the report, Waymo plans to improve how its vehicles detect construction zones and determine when they are already inside one. The update is also expected to improve how the system behaves once a construction zone is detected and to add additional operational protocols.

The issue underscores a persistent technical challenge in autonomous driving: prioritization. These systems constantly evaluate multiple potential hazards – other vehicles, road debris, lane markings – and must decide which to respond to first. In the cases outlined by regulators, the system appears to have given higher priority to certain roadway risks while failing to fully account for construction-related restrictions such as closed ramps.

Competition in the robotaxi sector continues to intensify. Uber Technologies, Lucid Group and Nuro recently announced plans to launch a robotaxi service in Houston next year, while Mobileye has said it is targeting a 2027 rollout. As companies move toward broader deployment, the ability to handle unpredictable, temporary road conditions is becoming a critical benchmark.