Not my normal news item, but this has general interest for those interested in the latest developments in autonomous driving. Maybe in my lifetime…
NPS 500 Autonomous Sensing Platform, the Most Compact, AI-Powered, Ultra Long-Range System that Intelligently Fuses and Integrates LiDAR, Radar and Cameras
Disruptive New Technology Establishes New Standard for Transportation Safety –
New High Performance LiDAR of More than 500 Meter Range and Innovative Radar Technology to See Around Corners
Neural Propulsion Systems (NPS), a pioneer in autonomous sensing platforms, today emerged from stealth to launch NPS 500™, the safest and most reliable platform for autonomous vehicles that enables the industry to reach Zero Accidents Vision. NPS 500 is the world’s first all-in-one deeply integrated multi-model sensor system focused on Level 4/5 autonomy.
The radically new sensor-fused system precisely interconnects the NPS revolutionary solid-state MIMO LiDAR, super resolution SWAM radar and cameras to cooperatively detect and process 360° high resolution data giving vehicles the ability to prevent all accidents. The densely integrated sensor system enables vehicles to see around corners and over 500 meters of range with ultra-resolution accuracy together with highly adaptive frame rate. The NPS 500 breakthrough capabilities make it 10X more reliable than currently announced sensor solutions.
“LiDAR, radar and cameras will all play significant roles in creating the ideal autonomous driving platform and there is no question that tightly connected sensors with onboard data fusion for automated driving enables more functionalities,” said Pierrick Boulay, senior analyst at Yole Développement. “This direction is unique and is likely to succeed in a market that could reach $25B in 2025* for sensing and computing in both ADAS and robot vehicles.”
“Our goal to prevent all transportation accidents is the holy grail for autonomous vehicles,” said Behrooz Rezvani, founder and CEO of NPS. “We are the sensing system behind the Zero Accidents Platform for large volume deployment at affordable cost. Existing technologies are not sufficient to achieve this paradigm, so we created our own more powerful LiDAR and radar. Our AI-driven sensor-fusion system processes this ultra-high resolution data to create the safest and most reliable solution in the market today. The NPS 500 slashes time-to-market for autonomous vehicle manufacturers, while being the most cost-effective.”
NPS 500 Product Details
The NPS next generation precision-built, multi-modal sensor system is the industry’s most advance autonomous driving solution that addresses physics-based limitations of each sensory system. The NPS 500 enhances and combines the strengths of LiDAR, radar and cameras to create a platform that leverages the capabilities of each technology, while addressing today’s challenges of Level 4/5 autonomy, including:
Cameras: Provide high resolution images, but lack depth information and depend on lighting conditions
Radar: Measure velocity with great precision, but have lower resolution than LiDAR and are vulnerable to interference from other radars
LiDAR: Provide super precision depth information, but its performance and reliability degrade in adversarial, weather and light conditions, and it can get occluded fairly easily
NPS 500 is the world’s first all-in-one deeply integrated multi-model sensor system focused on Level 4/5 autonomy.
Features:
LiDAR: Revolutionary new solid-state MIMO-LiDARTM architecture doubles range to ≥ 500 meters with super resolution and adaptive multi-beam search
Radar: New class of radar technology with 10X better detection reliability, simultaneous, multi-band 360° FoV, 70X better against other radar signal interference
Software: First ever AI fusion technology to “see-around-the-corner”
Chips: 650 Tb/s sensor-data processing on network of tightly connected custom signal processing chips
Benefits:
Range ≥ 500 meters @ 10% reflectivity
Doubling the reaction time currently available LiDAR
Significant increase in sensor data reliability
See-around-the-corner capabilities
Anticipating pedestrian’s movement well before reaching cross section
Detecting moving objects approaching intersections well in advance
Built-in redundancy for maximum reliability in harsh environments, bad driving and tough terrains
Low maintenance, automakers can efficiently rely on NPS sensors once the vehicles leave the dealership
Multi-beam adaptive scan up to 100 FPS to detect and track subtle movements
See thru occlusion
Reduced time to market
Cost effective
Low CAPEX and OPEX for OEM customers
Pricing & Availability
Customers and partners may contact [email protected] for more information.
About Neural Propulsion Systems (NPS)
The NPS mission is to eliminate all transportation accidents to achieve the Zero Accidents Vision. Founded in 2017 by Silicon Valley luminaries, including entrepreneur Behrooz Rezvani, former Ikanos and Quantenna founder and CEO. NPS is delivering the world’s first all-in-one deeply integrated multi-model sensor system for Level 4/5 autonomy for large volume deployment at an affordable cost. Its flagship product NPS 500™ precisely interconnects the NPS revolutionary solid-state MIMO LiDARTM, super resolution SWAMTM radar and cameras to cooperatively detect and process 360° high resolution data. The densely integrated deep sensor-fusion system gives vehicles the ability to see around corners and over 500 meters away, making it 10X more reliable than competitors.
My last business trip was to the preview of Hannover Fair 2020. This year, Hannover Fair is once again digital (virtual). Companies are beginning to send previews of their stands and demonstrations. This one is from Festo. This company has some creative engineers who have done some cool things with pneumatics. This description of their 2021 Hannover Fair stand follows.
The Productivity Master, a modular demonstration system for personalized USB memory sticks from Festo, shows how automation technology will evolve along the value chain when combined with digitalization. The Productivity Master demonstration will be the centerpiece of the Festo exhibit at the upcoming Hannover Fair Digital Edition April 12-16.
Thanks to seamless connectivity, everything fits perfectly, from the mechanical and electrical systems to the intelligence. The Festo automation platform provides an integrated and practical system to link all Festo engineering tools, components, and solutions in hardware and software.
The demonstration shows how a plant can achieve a balance between mass production and individualization of a finished product. The electrical products, the axis mechanics, the electrics, and the software are planned as a complete automation platform with seamless connectivity.
With this consistency, users save a lot of time in their machines and systems – from planning to commissioning – and gain process reliability because everything fits together – Industry 4.0 included.
Cloud-based concept
For example, the Productivity Master uses a cloud-based concept for registering and storing customer data complying with data protection regulations. This allows USB sticks to be produced with a personalized design as well as personalized data content. Customers could do this from home via the Internet without having to enter further data from other people on the supplier side.
Customers can use the same concept to reduce labor costs and coordinate the process from the home office. Festo’s IoT gateway securely connects the machine to the cloud so that operators cannot only retrieve production data from the cloud, but also access the machine’s diagnostic data, even if they are sitting hundreds of miles away in their home office.
Mask production and vaccines
The pandemic year 2020 revealed just such a contrast between mass production and individualization or small series production in the examples of mask production and vaccine research. The latter is more focused on precise data collection. For mass production, however, it is crucial to be able to use remote diagnostics and maintenance tools to maintain machine uptime as long as possible while protecting operator health.
“It doesn’t matter if the product weighs a few ounces or more than 200 pounds – being a technology-neutral company, Festo has the product range to do this,” explained Nigel Dawson, Head of Business Development Electric Automation at Festo. “While the Productivity Master features a mix of pneumatic and electric automation, from flow sensors to servo drives, from pneumatic slide units to electric Cartesian robots, it is the way these products are connected that enables true digitalization.”
Seamless connectivity
By leveraging industry-standard networks such as EtherCAT, OPC-UA, IO Link, and MQTT, Festo can create a seamless data stream from individual sensors to remote IO, servo drives, controllers, and the cloud. “This technology-neutral approach has a major impact on energy efficiency and sustainability. We call this ‘seamless connectivity,’” added Dawson.
The Productivity Master was designed in record time utilizing Festo’s state-of-the-art unified engineering environment which includes project planning capabilities and management of digital twins. This allows designers and developers to work collaboratively worldwide. With the Festo Handling Guide Online HGO, users select and dimension 2D or 3D gantries in just three steps. For quick and easy programming, configuration, and commissioning in a shared virtual environment, they use the Festo Automation Suite.
Digital Customer Journey
“Along the journey to digitalization, we enable customers to correctly and efficiently select and size products online, add them to a shopping cart in a seamless process, view pricing and delivery in real time, and track products throughout the supply chain,” said Dawson, describing this digital value chain. “With developments in artificial intelligence (AI), Festo is expanding concepts such as digital nameplates, digital twinning and, of course, AI-driven predictions of machine conditions for remote maintenance.”
Four fully connected stations
In a show setting, attendees create the production order at the first station by registering with their name and perhaps a picture. A QR code identifies the visitors at the machine. A vision sensor SBSI from Festo registers the visitor and starts the production order, followed by stock management. This mechatronic complete solution, made up among other things of a planar surface gantry EXCM in size 40 for fast picking and placing of the USB memory stick, offers all the options from stock management and workpiece handling to the cloud.
The USB memory stick is transported, printed, turned over, and then passed on. The work steps at the second station involve holding, moving, handling, rotating, positioning, and identifying the workpiece position as well as identifying, separating, gripping, and applying labels. Fast transfer to a label printer is carried out at station 3 by a highly dynamic linear gantry EXCT with dynamic servo motors EMMT-AS and servo drives CMMT-AS. The rotation functions are performed by electric rotary drives ERMH.
The process is completed in station 4, where files are uploaded to the USB memory stick. Station 4 operations include holding the workpiece, turning, and carefully positioning the cap and press-fitting delicate parts. This is done using the attractively priced electric cylinders EPCO and rotary drives ERMO.
2019-20: unprecedented events caused continuous negative growth for the first time for the industrial robotics sector
Swift recovery predicted: 9%+ growth forecast for 2021; CAGR of 4.6% predicted for 21-24
Uneven recovery for different regions and classes of robot – China to lead global revival
Semi-conductors, chemicals, and food & beverage industries to be major drivers of sector
We’ve seen the market report on robot sales from the trade association. Here is an independent analysis from London-based Interact Analysis. This new report on industrial robotics shows that the industrial robots sector is on track to deliver a swift recovery after a recent unprecedented two-year downturn.
During the economic slow-down in 2019 the industrial robotics sector saw shipments fall by 5.4%. The pandemic of 2020 saw a further fall of 5.9% as 80,000 fewer units were shipped than previously forecast. This was a difficult time for the sector, and it will take time to recover. Interact Analysis acknowledges that pre-COVID estimates of growth up to 2024 have now gone out of the window, but it has identified strong drivers for growth as economies emerge from the pandemic. The company’s in-depth research has concluded that new industry applications for robotics and more advanced technologies coupled with reducing prices will cause an acceleration in revenues in 2021, with an increase of 9.2% in revenue terms, and 9.6% in shipment terms. This will be followed by sustained growth up to 2024.
Articulated robots are predicted to be the slowest to recover to 2019 levels, given that the industry where they are predominantly used – automotive – has been hit so badly by the pandemic. Sales of SCARA robots, used in light duty pick and place and assembly operations are predicted to recover as their potential for use in different applications is recognised. The collaborative robot market, which saw negative growth for the first time in 2020, is forecast to bounce back with an impressive 15-20% year-on-year growth rate up to 2028.
The expected pattern of regional recovery for the industrial robot sector is familiar. The virus started in Asia, and then moved to Europe and North America, and the infection continues to spread. As a result, normal business operations will resume earlier in Asia, and this will impact on robot shipments in those areas, most notably in China, where the virus was brought under control by May 2020. The Chinese market is the only one forecast to surpass 2019 levels by 2021. This is in the main due to the large domestic demand coming from China itself, rather than an increase in China’s export markets which are, of course, still badly hit by COVID.
Jan Zhang, senior director at Interact Analysis says: “The downturn of 2019 and the shock of the pandemic tripped up the industrial robotics market, but it only stumbled. Factors such as staff shortages, social distancing, the ever-pressing need for efficiency and speed, and the realisation of the need to build in resilience in the face of world shocks such as COVID-19 mean that now more than ever, industries across the globe recognise the value of robotics. We believe the next few years will see the sector recover its footing and thrive.”
About the report
During the process of compiling this report, the Interact Analysis team conducted over 30 hours of interviews with 30 key industry personnel at leading robot companies and end-users. These were conducted face-to-face or by phone. Banks of data were collected and analysed, some direct from companies and held confidentially, some, such as company reports, from public sources. It must be emphasised that this report, more than any other, is an organic and evolving tool, offering a newly introduced product tracking service, ensuring users of the report have the latest data.
About Interact Analysis
Interact Analysis is an international provider of market research for the Intelligent Automation sector. Our team of experienced industry analysts delivers research into three core sectors: industrial automation, robotics and warehouse automation, and commercial vehicles. Intelligent Automation – which is the integration of artificial intelligence and automation – will change virtually every industry imaginable. This combination enables greater efficiencies, productivity, convenience, and scale. It has the potential to drastically alter the outlook for many traditional industries such as manufacturing, healthcare and automotive as well as to lead to the emergence of entirely new industries.
Network cybersecurity news updates from Indegy, now part of Tenable, and Nozomi’s partnering efforts.
A visitor to the show floor of the annual ARC Advisory Group Industry Forum in Orlando a few years ago might have been surprised at the sheer number of new cybersecurity companies exhibiting. Adding to this number were a couple of established companies who had drunk the kool-aid and established cybersecurity practices.
My first thought, having been down this worn path too many times in my career, centered on how these companies could survive and, indeed, even find enough market to grow. Many companies are formed to sell. I figured that most of these new cybersecurity company founders were as much looking for exits at the Forum as much as looking for customers. Since that time, many have been acquired.
We only hope that, in the long run, this burst of creativity in the field will improve industrial control system (ICS) cybersecurity. Although the recent incident at a Florida water treatment facility shows how far we still need to go.
Tenable Cybersecurity
I had a brief chat with Barak Perelman, VP of OT Security at Tenable and former CEO of Indegy to discuss the threats to cybersecurity opened up by the great dispersion of industrial workers due to the Covid pandemic.
Indegy was an operations technology (OT) firm brought into an IT company (Tenable) in late 2019. This is one way to bring IT and OT together. Perelman told me that overall IT and OT networks are more interconnected than ever. Threats can freely flow between networks. And 67% of OT organizations are reporting new and more sophisticated tactics being used against them.
I asked Perelman who within a prospective customer company Indegy had called on and who was the effective buyer. He said this was the biggest shift of the last two years. The customer was plant manager, engineering manager, etc. Now more likely the buyer is the IT cybersecurity team. The biggest success they’ve seen is with a combination of the forces, for example placing an engineer from the plant on the security team. Strategically, senior-level executives and the board level have become concerned. They went to their CISO first for results. So most Tenable projects are led by IT teams.
Another big change in the market is reflected in his slide deck. He used to have a slide on “air gap” but that has been dropped. “Everyone now understands that everything is connected.”
Prior to the pandemic and the movement to remote work in 2020, companies thought that they didn’t have exposure to remote access. Then they discovered that an engineer somewhere added an unauthorized cell modem for the ability to access the system remotely for troubleshooting purposes. After the pandemic, utilities, for example, have as much remote as possible. Many organizations understood they wouldn’t fight it anymore. If you can’t fight it, then join it.
Nozomi Partnerships
Companies forming partnerships has been one of the biggest trends in the market during the past couple of years. Here are a couple featuring cybersecurity company Nozomi.
Tempered Combines Strengths with Nozomi Networks
Tempered Networks and Nozomi Networks announced a new partnership and product integration to deliver a full-featured industrial security solution for network visibility, threat detection and remediation. The joint offering integrates Nozomi Networks’ leading network visibility, threat detection and incident response system with Tempered Network’s Zero Trust policy enforcement and centralized software-defined perimeter management console. Today’s sophisticated security threats are driving requirements for not only extreme visibility and intelligent threat detection, but also automated remediation that can lock-down vulnerable systems while ensuring continued availability for authorized access and continuity of business.
“As the leader in OT and IoT security visibility and threat detection, Nozomi Networks gives us a powerful partner to deliver greater insight and remediation capabilities to our customers,” said Jeff Hussey, Founder and CEO of Tempered Networks. “The AI-powered network analysis and anomaly detection that they provide can drive more accurate micro-segmentation and security policy enforcement into our Airwall platform, providing a more secure, rapid response approach against industrial-grade network attacks.”
“Tempered Airwall delivers the military-grade encryption and secure access policy enforcement that many of our joint customers rely upon to quickly remediate anomalies and threats in their networks,” said Chet Namboodri, Nozomi Networks Senior Vice President of Business Development and Alliances. “The combination of threat visibility and automated enforcement significantly improves security response. Ubiquitous threats like the SolarWinds attack continue to emerge and industrial connectivity for remote work and connected smart devices continue to accelerate. Our combined offerings provide strong detection and defense against the rapid proliferation of advanced persistent threats, actively buttoning down attack surfaces.”
The product integration includes the ability of Tempered to mirror secure traffic to Nozomi Networks solutions through a fully encrypted overlay for greater analysis and insight. Armed with AI-driven insights from Nozomi Networks, Airwall customers can take remediation steps or refine Tempered security policies through the Airwall Conductor management console API. The Nozomi Networks solution and Tempered Networks Conductor work in concert to refine Airwall zero trust policies and address identified threats, going beyond the capabilities of traditional network security devices like firewalls or remote access solutions.
“A two-way integration of network monitoring of IoT devices and secure, zero-trust, communications is brilliant,” said Richard Stiennon, industry analyst with IT-Harvest and author of Security Yearbook 2020. “Ensuring that all communications is stealthed and encrypted while preserving visibility into traffic is a winning combination.”
NanoLock Security and Nozomi Networks to Provide End-To-End Cyber Protection for Critical and Industrial Infrastructures
NanoLock Security, a leader in OT and IoT device-level protection and management and Nozomi Networks Inc, a leader in OT and IoT security and visibility, announced they have partnered to provide an end-to-end cyber protection and management solution to secure OT in critical and industrial infrastructure. The joint solution will be deployed in the Atlantica Cybernext Security Operations Center (SOC) to serve clients with the most technologically advanced solutions for protecting their business and infrastructure.
Nozomi Networks’ real-time OT and IoT security solution provides network visibility, threat detection, and operational insight for OT and IoT environments, while NanoLock’s device-level protection and management solution tackles the rapidly growing Advanced Persistent Threats (APT) from both outsider and insider adversaries.
Together, the joint security solution from NanoLock and Nozomi Networks introduces a holistic approach that spans the entire IoT and OT network. Adding NanoLock’s device-level protection and forensic data to Nozomi Networks’ advanced network visibility, threat, vulnerability and anomaly detection extends cybersecurity coverage to include:
Lightweight, passive cyberattack prevention for devices such as smart meters, data concentrators, and controllers, with near-zero processing, power requirements and memory footprint
Anomaly detection covering the network as well as IoT and OT devices
Unified alerts and deeper device-level as well as network-level forensic data
Centralized device visibility, risk monitoring, and intelligence management
Secured and enforced OTA (Over-The-Air) device updates
“With cybersecurity threats to critical infrastructure on the rise, our partnership with NanoLock Security delivers advanced, end-to-end protection,” said Chet Namboodri, Nozomi Networks Senior Vice President of Business Development and Alliances. “We’re teaming with NanoLock to strengthen utilities’ defenses against cyberattacks, using robust device and network-level detection and protection alongside extensive network visibility and risk assessment.”
“We’re delighted to partner with Nozomi Networks to introduce a broader IoT and OT cybersecurity solution,” said Yanir Laubshtein, NanoLock’s Vice President of Cyber Solutions. “We see a critical need for a cybersecurity solution that starts at the device level and spans the network to bring comprehensive detection, protection, and management. Our joint offering addresses that need, while also bringing operational efficiency analytics to critical and industrial infrastructure.”
As things are beginning to open again, I’m hearing of some more solutions for protecting employees from Covid infection. Recently I interviewed the Early on during the pandemic, I interviewed several executives who had solutions for preventing the spread of the SARS-CoV-2 virus as people either continued to work in manufacturing or would be called back at any time.
Founder/CEO of another such company–Soumi Eachempati, MD, FACS, FCCM, Co-Founder and CEO of Chelsea Health Solutions and Cleared4 and a former Professor of Surgery and Public Health at Weill Cornell Medical School.
Eachempati is sort of a “checklist guru” having formerly developed an app to be used in triage in emergency rooms when patients (or others) could answer a few quick questions on a tablet or phone to assist ER personnel expedite treatment.
Similarly, he maintains the manufacturing industry can protect their workers by implementing the following safety checklist:
· Identify, Test, Trace: The backbone of a solid identification strategy is a strong symptom monitor that identifies individuals who need to be tested and quarantined.
· Educate and Communicate: Educating workers about Covid-19 risks is as important as education on safety protocols and equipment use.
· Enforce Symptom Monitor Adoption: Less than absolute adoption is especially frightening when one considers that the individuals who do not participate may be the one most likely to harbor Covid-19 or its symptoms.
Back to the ER app, he had evolved the app into a flu checker. Then when Covid struck, and the app could be used to differentiate Covid from flu. Started Cleared4 to promote and sell the app.
Cleared4, a scalable, customizable HIPAA-compliant platform, helps organizations seamlessly monitor employee Covid-19 symptoms by quarantining those at risk and enabling low risk employees to return to school and work in keeping with the constantly changing state mandates. Cleared4 is a patent-pending health management platform that uniquely uses text messaging, health centric technologies, and on-site kiosks to obtain valuable health clearance for employees.
It is an SaaS application where companies (and others) can buy instances and are given private URL. Enter employees or other concerned people who need to be screened into the app via csv file. It’s a self-screener for now. Employee answers the questions on the app (also can input vital data like vaccines) and gets OK (or not) from the app. Obtains a personal QR code or the information is logged to the employee data in customer’s system. This allows entry into the building. Employees are tagged into geophysical groups. If someone reports Covid infection, app can be used to track who may have come into contact and notify them to monitor and test.
The platform is currently being used by at least one NBA and one NHL team, with more in negotiations.
Companies are also buying. A chemical company in Illinois, for example.
Could be available B2C for usual customers or even C2C say someone is giving a Bar Mitzva or arranging neighborhood playdates.
Knowing the number of times I’ve sat through safety videos at chemical plants, refineries, and the like and been logged in and given a card, I could imagine this app as part of the entire safety process. Great for contractors or vendors.
Drawback
The main drawback I saw in the app was that it depended upon the person to enter their own information. I asked a friend who works at a chemical plant about this. He told me they had done a turnaround late last year with about 900 contractors on site. They used an active monitoring application called Triax with few instances of infection during the work.
I guess I believe that some of each may not be a bad strategy. (By the way, the Triax people never contacted me. I’m just reporting a customer experience.)
I have to admit amusement when IT companies discover industrial technologies and approach them from what they know—huge databases, analytics, teams of data scientists, and the like. This vision system announcement could have described how I approached selling and installing vision systems in the mid-90s. (Obviously not enough of them—I switched careers to writing about the technology.)
However, many engineers and maintenance managers have learned about non-standard purchasing of automation components. AutomationDirect (back then, PLCDirect) broke the mold of needing a local distributor rep. Now searching eBay for parts is almost common. So, why not go to Amazon? Heck, we all have Prime, right?
This is interesting. I’d be very curious to find out how many of you are looking to Amazon not only for servers and Web Services, but also for components and devices.
Amazon Lookout for Visionuses AWS-trained computer vision models onimages and video streams to find anomalies and flaws in products or production processes
GE Healthcare, Amazon, and Basler among customers and partners using Lookout for Vision
Amazon Web Services Inc. (AWS), an Amazon.com Inc. company, announced the general availability of Amazon Lookout for Vision, a new service that analyzes images using computer vision and sophisticated machine learning capabilities to spot product or process defects and anomalies in manufactured products. By employing a machine learning technique called “few-shot learning,” Amazon Lookout for Vision is able to train a model for a customer using as few as 30 baseline images.
Customers can get started quickly using Amazon Lookout for Vision to detect manufacturing and production defects (e.g. cracks, dents, incorrect color, irregular shape, etc.) in their products and prevent those costly errors from progressing down the operational line and from ever reaching customers. Together with Amazon Lookout for Equipment, Amazon Monitron, and AWS Panorama, Amazon Lookout for Vision provides industrial and manufacturing customers with the most comprehensive suite of cloud-to-edge industrial machine learning services available. With Amazon Lookout for Vision, there is no up-front commitment or minimum fee, and customers pay by the hour for their actual usage to train the model and detect anomalies or defects using the service.
In today’s manufacturing industry, production line shutdowns due to missed defects or quality inconsistencies can result in millions of dollars of cost overruns and lost revenue every year. To avoid these expensive issues, industrial companies must maintain constant diligence to ensure quality control. Quality assurance in industrial processes typically requires human inspection, which can be tedious and inconsistent at best, or at worst, infeasible.
Computer vision brings the speed and accuracy needed to identify defects consistently; however, implementing traditional computer vision solutions can be complex. Building computer vision models from scratch requires large amounts of carefully labeled images for each element of the manufacturing process. Then, teams of data scientists need to build, train, deploy, monitor, and fine tune computer vision models to analyze each individual phase of the product inspection process. Even small changes in the manufacturing process (e.g. replacing an out of stock component with an equivalent alternative, updates to the product specifications, or a change in lighting) means having to retrain and redeploy the individual model and perhaps other models downstream in the production process, which is tedious, complex, and time consuming. Because of these barriers, computer vision-powered visual anomaly systems remain out of reach for the vast majority of companies.
Amazon Lookout for Vision offers customers a highly accurate, low-cost anomaly detection solution that uses computer vision to process thousands of images an hour to spot defects and anomalies – with no machine learning experience required. Customers send camera images to Amazon Lookout for Vision in real-time to identify anomalies, such as damage to a product’s surface, missing components, and other irregularities in production lines. Utilizing a machine learning technique called few-shot learning (where the machine learning model is able to classify data based on a very small amount of training data), the service needs as few as 30 images of the acceptable and anomalous state as a baseline to begin assessing machine parts or manufactured products.
In addition to enabling the service to detect anomalies without large amounts of training data, this capability also allows the service to be adaptable to a wide range of inspection tasks within industrial settings. After analyzing the data, Amazon Lookout for Vision then reports images that differ from the baseline via the service dashboard or the “DetectAnomalies” real-time API so that appropriate action can be taken. Amazon Lookout for Vision is sophisticated enough to maintain high accuracy with variances in camera angle, pose, and lighting arising from changes in work environments. Customers also have the ability to provide feedback on the results (e.g. whether a prediction correctly identified an anomaly or not), and Lookout for Vision will automatically retrain the underlying model so that the service continuously improves. This feature allows the technology to adapt to changes in the manufacturing process and even understand when variations are permissible or not based on customer feedback. This means that customers can be more nimble and adapt their processes based on competitive advantages or external factors impacting their operations.
“Whether a customer is placing toppings on a frozen pizza or manufacturing finely calibrated parts for an airplane, what we’ve heard unequivocally is that guaranteeing only high-quality products reach end-users is fundamental to their business. While this may seem obvious, ensuring such quality control in industrial pipelines can in fact be very challenging,” said Swami Sivasubramanian, Vice President of Amazon Machine Learning for AWS. “We’re excited to deliver Amazon Lookout for Vision to customers of all sizes and across all industries to help them quickly and cost effectively detect defects at scale to save time and money while maintaining the quality their consumers rely on – with no machine learning experience required.”
Lookout for Vision is available directly via the AWS console as well through supporting partners to help customers embed computer vision into existing operating systems within their facilities. The service is also compatible with AWS CloudFormation. Lookout for Vision is available today in US East (N. Virginia), US East (Ohio), US West (Oregon), EU (Ireland), EU (Frankfort), Asia Pacific (Tokyo), and Asia Pacific (Seoul), with availability in additional regions in the coming months.
GE Healthcare is a leading global medical technology and digital solutions innovator that enables clinicians to make faster, more informed decisions through intelligent devices, data analytics, applications, and services. “We are excited about the encouraging early results from Amazon Lookout for Vision that will promise to help improve the speed, consistency, and accuracy of detecting product defects across our factories,” said Kozaburo Fujimoto, Operating Officer, General Manager, Manufacturing Division, Plant Manager, GE Healthcare Japan. “As one of the world’s most trusted healthcare companies with more than a century of technological progress and digital innovations, we look forward to capitalizing on the benefits that AWS’s industrial machine learning services will potentially bring to our manufacturing environments.”
Amazon’s Print-On-Demand (POD) facilities print books on demand to fulfill customer orders. “With POD, since books are manufactured when ordered by a customer, it is imperative to ensure precision at every step of the manufacturing process to offer a fast delivery time and the highest quality books to our customers,” said David Symonds, Worldwide Director of POD for Amazon. “With Amazon Lookout for Vision, we can automate and scale visual inspection at each step of manufacturing while running at full processing speeds, helping us ensure a great customer experience.”
Basler is a global manufacturer and solution provider in industrial vision, providing cameras and machine vision systems for applications such as semiconductor inspection, robotics, food inspection, postal sorting, and inspections of printed images. “Reducing faults is one of the most important KPIs to consider for manufacturing companies. Traditional manual inspection is labor intensive and difficult to scale. By using computer vision for quality inspection, this process can be automated and lead to a significant reduction of costs,” said Gerrit Fischer, Head of Marketing for Basler AG. “Basler and Amazon Lookout for Vision provide a very lean architecture to adopt vision based anomaly detection in any manufacturing site. We’re excited to jointly provide our customers with complete vision solutions by combining Basler’s expertise in industrial vision and edge platforms with AWS’s investments in industrial machine learning.”
Dafgards is a household name in Sweden, manufacturing a broad assortment of foods. “We previously tried Amazon Lookout for Vision to automate the inspection of our pizza production lines to detect whether pizzas had enough cheese and the correct toppings, with good results,” said Fredrik Dafgård, Head of Operational Excellence & Industrial IoT for Dafgards. “We’re excited to extend Lookout for Vision to our other production lines such as hamburgers and quiches, to help us detect any anomalies like incorrect ingredients. Over time, we plan to scale Lookout for Vision across multiple production lines. Amazon Lookout for Vision will allow Dafgards to improve the consistency and accuracy of detecting defects and anomalies, allowing us to improve our overall production quality at scale.”
About Amazon Web Services
For almost 15 years, Amazon Web Services has been the world’s most comprehensive and broadly adopted cloud platform. AWS has been continually expanding its services to support virtually any cloud workload, and it now has more than 200 fully featured services for compute, storage, databases, networking, analytics, machine learning and artificial intelligence (AI), Internet of Things (IoT), mobile, security, hybrid, virtual and augmented reality (VR and AR), media, and application development, deployment, and management from 77 Availability Zones (AZs) within 24 geographic regions, with announced plans for 18 more Availability Zones and six more AWS Regions in Australia, India, Indonesia, Japan, Spain, and Switzerland. Millions of customers—including the fastest-growing startups, largest enterprises, and leading government agencies—trust AWS to power their infrastructure, become more agile, and lower costs.
