Waymo Reveals Custom Chip Powering 4,000 Robotaxis Across U.S. Cities

Waymo lifts the lid on the 'brain' powering its robotaxis

What's in the trunk of your car? Shopping bags? A spare tire? Jumper cables? What about a high-powered computer capable of performing up to one quadrillion operations a second? Only if you're Waymo. For the first time, Waymo revealed key details about the heavy compute "brain" housed in the trunk of its robotaxis, including chip architecture, processor specs, and internal component details. The details, published in a blog post today, also includes a list of hardware suppliers that Waymo uses to build the computers that power its driverless fleet. It's a revealing look at the world's leading player in driverless technology. Waymo is currently operating around 4,000 vehicles in over 10 cities, conducting approximately 500,000 paid trips a week. The company claims that its technology significantly reduces traffic crashes and injuries as compared to human drivers. And while a lot of the work to run its vehicles relies on heavy training and simulation that runs in the cloud, a good portion also takes place in the vehicles themselves. Waymo's robotaxis, festooned with cameras, lidar, and radar, take in massive streams of raw sensor data. That data is processed by artificial intelligence to instantly understand what's happening around the car, turning that information into safe driving commands in milliseconds. Its computers, which are installed in the trunks of its vehicles, are built to act like its eyes, ears, and brain, to process a full 360-degree view and make split-second decisions without any human backup. "Compute systems for autonomous driving handle highly diverse workloads," Satish Jeyachandran, VP of Engineering, and Daniel Rosenband, Compute Lead, write in the blog post. "At Waymo, we are designing a state-of-the-art system that would be considered impressive for a data center, with the added complexity of an in-vehicle operating domain and real-time requirements." Jeyachandran and Rosenband describe three core principles guiding the design of its compute: responsive, ruggedized, and redundant. The stack needs to be able to respond to sensor data as quickly and with as little latency as possible. This requires massive amounts of compute power, which Waymo has scaled 20 times in the past eight years. It also needs to be able to handle all the bumps and cracks in the road, as well as extreme temperatures. And it basically needs to be able to run two systems in parallel, in the event that one fails. These three principles are "non-negotiable," the executives write. It also needs to not take up the entire trunk. After all, there are passengers and their luggage to consider. And it needs to run nearly silent, so as not to compromise the rider experience. And while Waymo builds its compute stack itself, it also relies on a network of third-party suppliers to provide the components it can't build itself. Waymo's effort to explain how its multi-sensor system can fuse data in real time is especially noteworthy considering Tesla CEO Elon Musk's dismissal of these systems as inherently faulty. Musk has called lidar a "crutch" and "a fool's errand," arguing that sensor fusion between cameras, radar, and lidar introduces dangerous "sensor contention." He also contends that when differing sensor data disagree, figuring out which signal to trust creates deadly ambiguity rather than safety. But Waymo argues that its robust and redundant system is what enables it to deploy its vehicles at scale. "We built an [machine learning]-primary architecture to run advanced neural networks at minimal latency," the executives write. "To manage critical non-ML tasks like orchestration, data movement, and logging while maximizing time for ML computation, we pair our ML technologies with the best CPUs, GPUs, and accelerators. The result is a balanced, heterogeneous system." In the blog post, Waymo introduces its custom silicon chip, a 5-nanometer ASIC, built to handle the massive firehose of incoming data from its sensors. Instead of sending raw, messy data straight to the car's main brain, the chip sits at the front lines cleaning up the data, combining the sensor inputs, and running fast AI checks right as the information comes in. The chip's dedicated front-end processing provides 1,000 TOPS (trillion operations a second). And because it was built specifically for Waymo's autonomous driving operations, it fits neatly in the overall stack, helping keep reaction times minimal. Consider this scene below. Waymo's executives claim its system can more quickly process high-fidelity data from its 13 high-res cameras "simultaneously and in real-time" to deliver improved low-light perception. In that way, it can see a lot more stuff in the dark (the image on the right) that traditional cameras often miss (the image on the left). Jeyachandran and Rosenband conclude the blog post by running through a list of hardware suppliers who are helping build Waymo's cutting edge system: AMD, Micron, Nvidia, Samsung, SanDisk, Socionext, and TSMC. Rather than try to build its own compute system from scratch, the company is instead focusing its internal efforts on real-time sensor fusion and front-end machine learning, while leaning on the global supply chain for everything else. It's a revealing glimpse of the technology that Waymo uses to power its fleet of robotaxis. But the company doesn't get into one important aspect: How much does all this cost? Some estimates put it at $20,000 to $25,000 per vehicle for its sixth-generation hardware. While that's a huge decrease from the fifth-generation's estimated cost of $100,000 to $125,0000 per vehicle, it still shows how Waymo's continued expansion could be hampered by high hardware expenses. Building a brain to run an entire fleet of driverless cars is a monumental achievement, no question. But building a business that can make a profit is another challenge altogether.

Waymo has designed a robocar chip to stay ahead of Tesla

To reach their destinations safely, autonomous vehicles have just milliseconds to ingest and process streaming data from more than a dozen cameras. It's a job that's been handled with off-the-shelf AI components thus far, but Waymo has begun rolling its own AI ASICs to optimize the process. It's not alone. Revealed in a blog post Thursday, the Alphabet-backed robo-taxi startup's first custom silicon is designed to convert raw sensor data into driver responses as quickly as possible. Built on Taiwanese foundry giant TSMC's 5 nm process tech, the chip is specifically optimized to run both more traditional machine learning algorithms like convolutional neural networks and modern transformer models similar to those used to run AI chatbots or image generation models. According to Waymo, the chip's design incorporates more than 200 million miles worth of autonomous driving data, and is tuned to maximize responsiveness, reliability, and redundancy. Prior to this, Waymo had employed Intel FPGAs for sensor processing. FPGAs are ideal in low latency applications, which is one of the reasons why high frequency trading often takes place on them. However, compared to dedicated silicon, FPGAs are notoriously difficult to program for and lack the compute density achievable using application specific hardware. Accidents can unfold in a fraction of a second, far too quickly for a remote operator to take over. So Waymo designed the chip with a major focus on minimizing latency. "Within those critical milliseconds, advanced ML models build a high-fidelity understanding of the environment to evaluate the safest path forward," the company explained. This includes performing temporal noise reduction to improve low light visibility in real time. All of that requires a considerable amount of computation. The robo-taxi startup claims its ASICs are capable of churning out more than 1,000 TOPS of AI performance. But without knowing the precision and power levels the chips are operating at, it's difficult to draw comparisons to existing autonomous vehicle and robotics platforms. We've reached out seeking clarification, and will let you know if we hear back. But considering that Waymo is specifically advertising TOPS, we're probably looking at INT8 performance, which would put it in the same ballpark as Nvidia's Drive AGX Thor platform. In addition to offering plenty of compute, the chip also needs to be reliable. Vehicles are exposed to a near-constant stream of vibrations, shock, and extreme temperature swings, unlike anything you'd see in a datacenter. In order to combat this, Waymo has employed multiple layers of redundancy. The chips themselves are liquid cooled by the same coolant system used by the vehicle itself, ensuring that the silicon maintains optimal temps regardless of the weather. Meanwhile, to ensure a hardware fault doesn't put passengers at risk, Waymo says each vehicle is equipped with a pair of ASICs. Under normal conditions, the chips behave as a single unit. But in the event one fails or produces an erroneous result, the other can take over. Air and space craft often include a third system to serve as a tiebreaker in the event of an upset. So, it's not surprising to see this level of redundancy employed for a vehicle that's going to be hurtling down roads filled with other vehicles, pedestrians, and obstacles. To be clear, Waymo's custom ASIC isn't responsible for all of the vehicle's functions. The company says that it is working on several other custom chips and systems, but for now, non-ML tasks like orchestration, data movement, and logging are handled by components provided by its partners, which include AMD, Micron, Samsung, Sandisk, and Nvidia. Waymo is far from the only autonomous vehicle vendor rolling its own custom silicon. Tesla, which launched a limited Robotaxi service in Austin after years of missed deadlines, has been developing custom chips for its vehicles for years. Waymo will share more detail on its ML accelerators next week during the annual Hot Chips conference at Stanford. ®

Waymo's custom chip decides what the car sees

Waymo has published the architecture of the computer in its robotaxis, a custom 5nm chip it designed, and the names of its seven hardware suppliers. Nvidia and AMD are still two of them. The chip handles sensor data before the driving system sees it. Waymo has published what is in the boot of its robotaxis. A blog post on Thursday set out the architecture of its onboard computer, introduced a custom 5-nanometer chip of its own design, and named the seven suppliers it builds the system with. The company calls it the first look under its trunk. For a firm that discloses very little about its hardware, it is a substantial opening. What the chip actually does This is where the coverage and the company diverge, so it is worth being precise. The part is an application-specific integrated circuit built on a 5nm process. Waymo describes it as handling "the massive influx of raw data before it reaches our core ML brain". It extracts information from raw lidar, radar and camera streams, including temporal denoising for better perception in low light, then feeds that into a separate inference engine that runs the sensor-fusion models. In other words, it is a front end. It prepares what the driving system sees rather than deciding what the car does. That matters for the number everyone quoted. Waymo says "these ASICs alone deliver over 1,000 TOPS of ML performance dedicated to front-end processing and ML models". Bloomberg reported that on that measure Waymo would land level with Nvidia's latest autonomous-driving systems. That compares one component doing one job against complete driving platforms. It is not a claim Waymo made. Waymo has not dropped Nvidia Nor has it dropped AMD. Both appear in the supplier list Waymo published, alongside Micron, Samsung, Sandisk, Socionext and TSMC. The company describes a "balanced, heterogeneous system" that pairs its own machine-learning silicon with what it calls the best CPUs, GPUs and accelerators available. The custom chip adds a capability. It does not replace the merchant parts around it. Waymo also says the ASIC is "just one of several exciting custom components we're developing". More in-house silicon is coming, and the company has not said what. Three constraints, and one of them is a safety design Waymo builds the system around three requirements it calls responsive, ruggedised and redundant. The first is latency. It has coined the phrase "pixels-to-actuation" for the delay between a photon hitting a sensor and the car acting. Raw compute has scaled 20 times in eight years to shorten it. The second is physical. The computer sits in a car, not a rack, so it endures vibration, shock and temperature swings from Midwest winters to Phoenix summers. Waymo plumbs it into the vehicle's liquid cooling system, and it has to run silently and leave room for luggage. The third is the interesting one. "Our compute is designed like two independent engines," the post says. They normally run full parallel workloads as a single unit, and if one develops a fault the other takes over. There is no human backup, so the redundancy is in the hardware. The scale, and whose numbers they are Waymo cites more than 200 million miles of fully autonomous experience behind the design. It runs about 4,000 vehicles in more than 10 cities and completes roughly 500,000 paid trips a week, The Verge reported. Every one of those figures comes from Waymo. The company also says its technology reduces crashes and injuries against human drivers, on its own safety data. No outside body has audited any of it. Where the silicon is going The system processes 13 high-resolution cameras at once. That count matches the Ojai, the purpose-built four-seater that also carries four lidar sensors. Zeekr builds the Ojai, and Zeekr is an arm of the Chinese carmaker Geely. TNW covered the vehicle in May as cheaper to build, harder to break, and made in China. Bloomberg reports it is now being equipped with the custom chip, though nobody has said how many are fitted or by when. The politics are live. Washington has spent the year tightening rules on Chinese vehicle hardware, and one supplier the Pentagon calls a military company still has Chinese lidar running in American robotaxis. Waymo designing the brain while Geely builds the body is one response to that exposure. Neither company put it that way. A day before the chip post, Waymo opened the Ojai to all riders in Los Angeles, Phoenix and San Francisco, Engadget reported. It had been invitation-only. The pattern, and the node Waymo is not acting alone. Its parent has been designing silicon for years to hold down AI infrastructure costs, and on Wednesday Marvell handed Google a $12.2bn share option in a custom-chip arrangement. One comparison is worth sitting with. Waymo chose 5nm, a mature node rather than the leading edge. BYD has built a 4nm driving chip for a car costing $10,000. Node size is not capability and the two parts do different work, but it is not the ranking most readers would assume. Socionext is the name to notice in that supplier list. The Japanese firm designs custom ASICs for other companies rather than selling its own, which is the sort of partner a company turns to when it wants silicon of its own without building a chip division from scratch. What the post does not answer Publishing a supplier list is not normal corporate behaviour, and the timing invites a question. Waymo faces a rival that talks about its hardware constantly, and Chinese competitors expanding abroad through partners. Specifications are an argument. What the post does not address is the year Waymo has had. It issued its sixth recall after robotaxis drove into highway construction zones 13 times, suspended freeway service in May and restored it in July. Faster front-end processing does not fix a decision made downstream of it, and Waymo does not claim otherwise. The engineers will present more at Hot Chips. That is where the specifications get questioned by people equipped to question them.

Waymo details custom AI silicon for autonomous vehicles

The company said its 'purpose-built 5nm ASIC' chip aims to optimise AV systems for 'low-latency performance'. Waymo, the Alphabet-owned autonomous vehicle (AV) manufacturer, has built a custom AI chip to improve the future performance of its robotaxis. In a blogpost yesterday (20 August) written jointly by the company's vice-president of engineering and its compute lead, the company said its "purpose-built 5nm ASIC" chip aims to optimise AV systems for "low-latency performance", leading to better and faster execution of real-time driving commands. Waymo said its AV systems integrate co-designed hardware, sensors and algorithms to "process, fuse and run advanced neural networks on raw sensor data in real time", and that its custom chip is built to "handle the massive influx of raw data" that an AV system must process. Because Waymo's system "handles the entire task of driving without a human backup", the blogpost read, the company's AV compute is built around three "non-negotiable" pillars, one of which is responsiveness - so the system is capable of "constantly processing decisions within milliseconds" to make safe decisions with "ultra-low latency". The ASIC chip can deliver more than 1,000 TOPS - trillions of operations per second - of machine-learning (ML) performance "dedicated to front-end processing and ML models", according to Waymo. The company said that aside from developing its custom silicon, it is also partnering with companies such as AMD, Micron, Nvidia, Samsung, Sandisk, Socionext and TSMC to scale Waymo's technologies. The company currently serves riders in 11 US cities, according to its website, and it has plans to expand to nearly 20 more, as well as London and Tokyo. In June, a recall notice showed that Waymo was pulling nearly 3,900 robotaxis from US streets over a software issue that let AVs enter and drive in closed freeway construction zones. In May, the company had to recall nearly 3,800 robotaxis from US cities over a software issue that could allow vehicles to drive onto flooded roadways. Waymo raised $16bn in a February funding round led by Dragoneer Investment Group, DST Global and Sequoia Capital - with more investment from parent company Alphabet and several others - that put the AV company's valuation at $126bn. Earlier this week, a three-way collaboration between Uber, Pony AI and Verne began offering robotaxi services to Uber customers in Zagreb, Croatia. Don't miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic's digest of need-to-know sci-tech news.

Waymo details the custom chip in its autonomous driving system

Waymo details the custom chip in its autonomous driving system Waymo LLC today shared new details about the computing module that powers its autonomous taxis. The Alphabet Inc. unit operates about 4,000 vehicles in 11 U.S. cities. Most are based on the Jaguar I-Pace crossover, while the rest are Zeeker minivans and Hyundai Ioniq 5 SUVs. Consumers order rides via a standalone app and Uber. Waymo's autonomous driving module is in its sixth iteration. Much of the current version's computing power comes from two custom ASICs, or application-specific integrated circuits. An ASIC is a chip designed from the ground up for a specific set of tasks. Waymo says that the accelerators can perform more than 1,000 trillion calculations per second. "To make real-time driving decisions, the autonomous system operates entirely onboard, constantly processing decisions within milliseconds," Waymo executives Satish Jeyachandran and Daniel Rosenband wrote in a blog post. "We have engineered our stack for ultra-low latency, minimizing the delay from first pixel to action." Waymo makes its ASICs using Taiwan Semiconductor Manufacturing Co.'s five-nanometer node. Waymo's decision to use the technology over one of TSMC's newer processes may have to do with the fact that it's available in a vehicle-optimized edition. N5A, as the customized node is called, includes optimizations designed to reduce the risk of manufacturing faults. It complies with three different auto industry reliability standards. Waymo included two ASICs in its autonomous driving module to mitigate the impact of hardware failures. If one of the processors goes offline, the other can take over. The chips process data from a sensor suite that comprises 13 cameras and four lidar devices. Cameras are better at capturing fine-grained details such as text on street signs, while lidars work more reliably in low-light conditions. Waymo's sensor suite also includes six radars that collect data points such as the speed of nearby vehicles. The Alphabet unit optimized its ASICs for artificial intelligence inference. The chips run AI models optimized for sensor fusion, the task of combining readings from different types of sensors into a single dataset. The process involves adjusting the raw data for the fact that a car's sensors collect measurements from different angles. Sensor fusion models are also responsible for error correction. According to Waymo, its autonomous driving module uses an error correction method called temporal denoising. The technique enables AI models to verify that a frame contains noise by checking adjacent frames. Besides the ASICs, Waymo's autonomous driving modules also include central processing units and graphics processing units. They chips are stored in a ruggedized enclosure designed to operate in challenging conditions. The system dissipates heat with the help of the liquid cooling system in the host vehicle's battery. "We have engineered our compute to thrive in the physical world with remarkable endurance and reliability from the component to the system level," Jeyachandran and Rosenband wrote. "Our hardware operates under constant vibration, shock, and extreme temperatures."

Waymo Unveils First Custom Robotaxi Chip With More Than 1,000 TOPS of AI Processing Power - Alphabet (NAS

Alphabet Inc.'s (NASDAQ:GOOG)(NASDAQ:GOOGL) Waymo has revealed its first custom-designed robotaxi chip, a purpose-built 5-nanometer ASIC delivering more than 1,000 TOPS of machine-learning performance for front-end sensor processing in its newest Ojai vehicles. Custom Chip Handles Real-Time Sensor Processing Waymo described the silicon in a press release as "a specialized ML powerhouse engineered exclusively to process, fuse, and run advanced neural networks on raw sensor data in real time." The chip handles streams from lidar, radar and cameras before feeding processed information into the vehicle's core machine-learning system. The Alphabet unit said its latest system can process data from 13 high-resolution cameras simultaneously and in real time, including temporal denoising designed to improve low-light perception. Waymo has increased raw onboard compute power 20-fold over eight years. Waymo Still Relies On Outside Chipmakers The ASIC does not replace outside chipmakers. Waymo said it continues working with Advanced Micro Devices Inc. (NASDAQ:AMD), Micron Technology Inc. (NASDAQ:MU), Nvidia Corp. (NASDAQ:NVDA), Samsung Electronics Co. Ltd. (OTC:SSNLF), Sandisk Corporation (NASDAQ:SNDK), Socionext Inc. (OTC:SOCNF) and Taiwan Semiconductor Manufacturing Co. (NYSE:TSM). Nvidia's DRIVE AGX Thor can deliver up to 1,000 INT8 TOPS, although Waymo has not disclosed the precision behind its own TOPS figure and the architectures serve different workloads. Tech Nvidia Partner Says Tesla, Waymo Are Growing the Physical AI Market -- Not Competing With It Ouster CEO Angus Pacala says Tesla and Waymo are accelerating Physical AI adoption rather than competing with the lidar maker. 3 min read Read this article The approach contrasts with rival Tesla Inc. (NASDAQ:TSLA), which relies on cameras and AI software rather than Waymo's lidar-radar-camera sensor suite. Tesla also designs autonomy silicon and said this year that development of its AI5 and AI6 inference chips is progressing, with production targeted for 2027 and 2028. Reuters reported Tesla is preparing to launch its purpose-built Cybercab as the U.S. robotaxi race accelerates. Trending Get a 1% Match on Your First Deposit of $1,000+ Tesla's expansion has prompted investors to debate whether it could eventually challenge Waymo's lead. Ojai Rollout Expands Waymo's Robotaxi Lead Waymo's chip arrives alongside the rollout of the Zeekr-built Ojai, which uses its sixth-generation Driver. Waymo unveiled the updated Ojai in January and is scaling Waymo-enabled vehicle capacity toward tens of thousands annually at its Mesa, Arizona, factory. Waymo says its compute system runs two independent engines in parallel so one can take over if the other fails, a design it considers essential because no human driver is available as backup. The company now provides more than 500,000 fully autonomous electric-vehicle trips per week, a milestone that keeps it ahead of Tesla and Amazon-backed Zoox in U.S. robotaxi scale as both rivals expand. Tech Waymo Gets California Green Light for Robotaxi Expansion Across Bay Area and Los Angeles Alphabet's Waymo expands driverless ride-hailing across California, adding San Francisco, LA, Sacramento, and San Diego. 3 min read Read this article Photo courtesy: Shutterstock Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.