BookmarkSubscribeRSS Feed

AIoT Solution for connected Vehicles

Started ‎02-12-2021 by
Modified ‎10-20-2022 by
Views 2,462
Team Name  Nupeak Neurons
Track Manufacturing
Use Case Analytics for Connected Vehicles
Technology AI + IoT
Region India
Team lead Dipen Shah
Team members @Dipenshah @Kashishjoker @ajay_vemula  @Akash_Nupeak @akshay_solanki @Viral_Trivedi 

 

 

Business Landscape  

 

“A tsunami of digital innovation is disrupting our century-old transportation industry, as dream vehicles evolve to autonomous transport vessels and cities drown in traffic. The ability to deliver sustainable, digitally enabled transportation systems to cities – and shared, on-demand transportation experiences to citizens – is fundamentally changing the game in transportation.” 

 

The headlines and major company announcements share a common theme: Competitive disruption is reshaping business models and organizations’ very futures. Around the global automotive industry, component and original equipment manufacturers (OEMs) are taking a hard look at where their future growth will come from − and it’s not all based on their core businesses. New technology has opened the door for new services and revenue streams. 

The path many see is lined with connected services that represent new sources of revenue and inventive ways to better serve customers. Strategy shifts are taking manufacturers from fundamentally producing components and vehicles to holistically improving the lives of customers in multiple ways. In the near future, the automobile could represent the greatest mobile device available to consumers. 

Automakers and their partners see ripe opportunity in mobility, connectivity services and other product enhancements that will endear current and future customers. Connected and smart mobile technology for autos have been decades in the making. Now the commercial gains of these efforts are becoming near-term realities. 

 

Brief Background  

Dipenshah_8-1613053071609.png

 

From cars and trucks to factories and dealers, many automotive manufacturers and suppliers are already collecting information from the connected devices that send and receive data over the internet of things (IoT). Analysts project there will be more than 380 million connected vehicles on our roads by 2021. This opens exciting opportunities for the global automotive industry, prompting automakers to consider options for future success. However, traditional approaches to data management and analytics may not be sufficient for sustaining value in this new, connected world. 

Simply collecting data from connected sensors, systems or products is not enough. To benefit from the promise of IoT data, businesses need to be able to shift analytics from traditional data centers toward devices on the edge – the “things.” The challenges arise from the complexity – and risks – inherent in capturing and analyzing extreme volumes and varieties of the data torrents flowing from ever-increasing numbers of things. 

Today’s automotive manufacturers need more flexibility about where, when and how to manage and analyze IoT data. And they must understand which data is relevant, so they’ll know what to store and what to ignore. To get there, they need a trusted, automated solution. 

 

Executive Summary 

 

In the IoT, objects or sensors with embedded computing devices connect to the Internet to send and receive data. This behavior represents a significant architectural challenge because these devices generate enormous amounts of data, currently about exabytes per day. In addition, these new IoT-enabled devices produce many different types of data; this data is often very noisy, produced continuously, and uses a variety of protocols; and it all needs real-time analysis and response.  

Traditional computing models send the data to the core data center for analysis. However, this approach is impractical in many scenarios because of the volume of data being produced and the need for real-time analysis and response times measured in milliseconds. 

As a result, a new model for analyzing IoT data at the edge of the network has emerged. This model moves the analysis and response close to the devices that generate the data, reducing latency and also reducing the load on the network and the core data center. 

Our solution covers the full analytics life cycle, starting with data capture and integration and extending to analytics and deployment. 

 

“Partnerships are being forged between traditional manufacturers and IT giants to develop connected vehicle technologies.” 

 

Key Benefits 

 

In a fast-paced world, the insights we get from data have a shorter shelf life. They need quick action, or they lose value. By definition, mobility is motion, so mobility decisions must be made on data in motion. 

Manufacturers are rolling out remote diagnostics and automated, proactive maintenance programs. 

Smart apps are reinventing the way urban citizens get around. 

With SAS, below Key benefits can be achieved: 

Dipenshah_9-1613053091203.png

 

This is just a glimpse into the areas where analytics will propel connected vehicle and mobility initiatives, transforming the auto industry − and transportation at large. 

 

Business Use Cases  

 

From a customer experience standpoint,  

The connected vehicle offers tremendous potential for enhancing the entire customer journey with more valuable interactions at many points of contact. Whether it’s about proactive dealer notifications and support, on-board diagnostics that boost vehicle performance, or getting targeted and personalized offers while driving, connectivity can be a real boon to revenues, retention and loyalty. 

From a manufacturer standpoint,  

Connected vehicles are already having a huge influence on quality and reliability. For example, telematics data identifies defects that have made it into the fleet, which helps speed time to detection and minimizes the impact. The next frontier is predictive maintenance − reading raw sensor data (e.g., to translate diagnostic trouble codes) − to provide an alert to a likely future failure, so you can act fast to resolve the issue and preserve relationships with customers and Manufacturers.” 

Being connected through the Internet of Things – to send, receive and often act on data – results in many smart IoT things that we can use to build a more secure, convenient, productive and intelligent world. 

Already, Internet of Things capabilities play a significant role in businesses’ digital transformation efforts. When we combine IoT data with advanced analytics and AI – leading to the “Artificial Intelligence of Things” – the possibilities seem endless. 

Below listed possible Use Cases will enable Manufacturers to better understand the benefits: 

 

Dipenshah_10-1613053100498.png

 

Connected Vehicle Integration. Third-party application providers and systems that interact with the connected vehicle for accelerated connected vehicle integration. Provides improved insights into vehicle operation and ensures seamless information flow to the broader vehicle ecosystem, which can include dealers and vehicle manufacturers. Analysis's data in motion by processing huge volumes at very high rates (in the range of millions per second) – with extremely low latency (in milliseconds). You can also embed this powerful solution in devices to shift intelligence to the edge. 

Vehicle Health Management. Using data from connected vehicles, manufacturers can get a sense of parts that aren’t working as well as they should. They can anticipate when they need to find a new or better parts supplier, proactively eliminating consumer complaints before they even happen. They can alert customers to product recalls as soon as they get in the vehicle. They can better gauge anticipated warranty usage, establish smarter lengths of warranties, and build better vehicles from the start. 

Fine-Tune Repair Opportunities. Enables proactive service alerts and preventative maintenance diagnostics to help reduce warranty cost, improve product quality and build brand loyalty. Uses remote diagnostics and safety services to help prevent breakdowns and increase vehicle quality. 

Tracking & Tracing. The Journey Log information collection, processing and storing the vehicle trip information and makes it available for different services and applications like Place of Interest, Area of Interest, and Road based geofencing etc. 

Reporting & Dashboard. MIS Reports & Dashboard services enables Manufactures for better decision making based on below approach: 

 

  • Manufacturer Level Reporting – Vehicle details, performance tracking, diagnostics tracking, Dealers Services tracking & Parts Optimization reports etc. 
  • Vehicle Owner Level Reporting - Driving behavior analysis & driving violation alerts like over speeding, engine idling, harsh braking, high acceleration, trip by trip analysis etc. 

Smart Mobility Services. This includes providing Mobile Application to Vehicle owners to provide information about driving behavior and simplified UX. 

  • Login based Services -The driver must know his/her Company ID, Driver ID and Password to log in to the application. 
  • Security Impact Alerts – Immobilization, over speed alerts, Harsh Braking, Harsh Acceleration, SMS/Email alerts. 
  • Panic Button at distress – In the event that you find yourself in an unpleasant situation and wish you could let your friends and family know your location 
  • Push Notification Events – Any Manufacturers promotions and event notifications to all vehicle owners for upsell and cross sell. 
  • Two-way communication / Feedback – Vehicle owners can share feedback with Manufacturer from Mobile Application directly. 

 

Current Challenges  

Vehicles can produce upwards of 560 GB data per vehicle, per day. This deluge of data represents opportunities to derive value from the continuous stream of data and challenges in processing and analyzing data at this scale. 

The main challenges in developing a platform to connect and manage vehicle data include: 

  • Device management. To connect devices to any platform, you must be able to authenticate, authorize, push updates, configure, and monitor software. These services must scale to millions of devices and provide persistent availability. 
  • Data ingestion. The platform must reliably receive, process, and store messages. 
  • Data analytics. You can perform complex analysis of time-series data generated from devices to gain insights into events, tolerances, trends, and possible failures. 
  • Applications. Developers must create business-level application logic. This logic must integrate with existing data sources, including third-party sources and on-premises data centers. 
  • Analytical models. You must have predictive models based on current and historical data in order to predict business-level outcomes. 

To derive value from vehicle data, you must be able to ingest, store and process device data at scale. You must process data securely throughout the platform, and you must be able to scale processing, storage, and analytical applications to handle the amount of data generated from millions of devices in various geographies. You need rapid data analysis and advance predictive capabilities and feedback loops for applications from machine learning. 

SAS provides a robust platform for data ingestion, Internet of Things (IoT) device management, storage, analysis, and machine learning predictions. Centralized device management of a gateway per vehicle simplifies the control plane and data plane for sensors and data sources while helping to provide security and operational boundaries with cloud-based systems. 

 

 

Value Proposition - Intersection Capabilities  

 

Connected vehicle and mobility services are probably the most significant initiatives to hit the automotive industry since the invention of the assembly line – and may well have a more fundamental impact on the industry as a whole. 

SAS® Puts Mobility and Connected Vehicle Analytics into Action for Manufacturers as follows: 

Dipenshah_11-1613053163431.png

 

 

At the intersection of Connected Vehicles & Analytics  

 

There are a host of opportunities to support better decisions about Manufacturers’ internal operations and new ways to improve the customer experience:  

  • Quality and reliability. Automatically generate quality alerts at the individual or fleet level to spot issues sooner, reduce time to resolution and better manage warranty costs.  
  • Driver safety. Use driver behavioral data to spot aberrations and motivate better driving habits.  
  • Customer experience. Make tailored and relevant offers to customers for fuel, parking, restaurants and more based on location, direction of travel, time of day and customer preferences.  
  • Dealer services. Provide predictive maintenance alerts and optimize the availability of parts, technicians and bays to service customers’ vehicles.  
  • Usage-based insurance. Analyses real-time driving behavior to predict accident risks and offer usage-based insurance for car-sharing, ride-sharing or on-demand transportation services. 
    • Infotainment. Monetize drive time with customized content.  
    • Behavior/policy monitoring. Monitor young drivers for unusual deviations, shut off and recover stolen vehicles, and ensure that rental vehicles are not traveling out of range. 

At the intersection of Analytics & Mobility  

 

Mobility analytics enables transportation planners to make smarter decisions about the flow of people, brings new convenience and transportation options to consumers, and enables whole new categories of revenue services. These examples from early movers only hint to future possibilities:  

  • New mobility services. Support dynamic car sharing, ride sharing, bike sharing, and multimodal transportation offers, and pay-as-you-go models. This includes giving consumers up-to-the minute information about smart parking opportunities and arrival/departure details for other transportation options.  
  • Tailored customer experience. Build better customer experience models and make the best offers to consumers based on their location, time and other factors.  
  • Situation management. Optimize city flow and emergency response models, provide real-time road conditions, alerts and suggested alternate routes. 

 

Solution Landscape 

 

The connected vehicle solution includes capabilities for local computing within vehicles, sophisticated event rules, and data processing and storage.  

The solution is designed to provide a framework for connected vehicle services, allowing you to focus on extending the solution's functionality rather than managing the underlying infrastructure operations.  

You can build upon this framework to address a variety of use cases such as voice interaction, navigation and other location-based services, remote vehicle diagnostics and health monitoring, predictive analytics and required maintenance alerts, media streaming services, vehicle safety and security services, head unit applications, and mobile applications. 

 

Solution Components 

 

The diagram below presents the components and features SAS can build using the solution Components: 

Dipenshah_12-1613053213542.png

 

 

 

Azure Sphere 

 

With Azure Sphere, manufacturers and solution providers can build devices that are secure from the start and that stay secure throughout the device lifetime. 

A connected vehicle is equipped with Internet access, often including a wireless local area network to share data with devices both inside and outside the vehicle. Connected vehicle features are emerging in safety, navigation, infotainment, diagnostics/efficiency and payments. 

Vehicle Sensors. The device has a pre-installed SIM card through which it sends all the data it gets from Vehicle's engine and its inbuilt sensors to our server where we analyze the data and provide you with the relevant information.  All wireless data transmission is encrypted using bank level security algorithms to ensure that all your data is protected and secure.  

 

Sensor data 

Type of network 

Application 

Vibration sensors (ring and triaxial accelerometer) 

ANN 

Fault Diagnosis of rolling element bearings 

Vibration sensors 

ANN 

Fault diagnosis and machine condition monitoring of induction motors 

Frequency spectrum extracted from vibration sensor data 

2D CNN 

Fault detection in rotating machines and gearbox 

Vibration data from Input and output shaft 

2D CNN 

Small fault diagnosis 

Vibration sensor data 

1D CNN 

Fault diagnosis 

Frequency spectrum of Vibration sensor data 

2D CNN 

RUL Estimation 

Vibration sensor data 

1D CNN 

Structural damage detection 

Multimodal data with 58 sensors on the engine such as temperature and pressure etc. 

LSTM—RNN 

Fault diagnosis and RUL estimation of aero engines 

Frequency spectra from vibration sensors 

LSTM—RNN 

RUL Estimation of rotating machinery 

Dynamometer, accelerometer and acoustic sensors 

LSTM—RNN 

Forecasting and prognosis of milling machines 

Multimodal data with ambient temperature sensor, motor temperature sensor, load and current data 

Auto encoder 

Fault diagnosis 

Acoustic sensor data 

Auto encoder 

Fault diagnosis in rolling bearings & Anomaly detection in air compressors 

Vibration signals 

Auto encoder 

Motor fault classification 

Temperature sensors 

Auto encoder 

Anomaly detection in gas turbines 

Multimodal sensor data from a turbofan engine 

LSTM based Auto encoder 

Sensor data forecasting for RUL 

 

 

IoT Hub Azure IoT Hub provides a cloud-hosted solution back end to connect virtually any device. Uses device-to-cloud telemetry data to understand the state of your devices and define message routes to other Azure servicesIn cloud-to-device messages, reliably send commands and notifications to your connected devices and track message delivery with acknowledgment receipts. Automatically resend device messages as needed to accommodate intermittent connectivity. 

 

Connected Vehicle Communication Layer 

 

Azure Streaming. Each low-level message from each vehicle must be transmitted, processed, and then stored for later processing and analytics. The messages are streamed at the end of each driving session or in real-time using End-to-end analytics pipeline that is production-ready in minutes with familiar SQL syntax and extensible with JavaScript and C# custom code. 

Pub-Sub Engine. Cloud IoT Core's protocol bridge provides communication with the vehicle devices using MQTT. After the vehicle endpoints are authenticated, the protocol bridge accepts messages and forwards each to Cloud Pub/Sub. Cloud Pub/Sub is a globally scalable message queuing system, making it an excellent choice to handle the streams of vehicle data while at the same time decoupling the specifics of the backend processing implementation. 

 

Data Processing & Storage 

 

ETL Processing. As data volume grows then need of data transformation, enrichment and then store telemetry data becomes essentialData Processing Engine is integrated into solution components for faster processing & storage.  

Structured DB. A scalable NoSQL database service with consistent low latency and high throughput, making it an ideal choice for storing and processing time-series vehicle data. The enriched, raw device data is initially stored in Cloud based Data Storage (on-prem is also feasible) for later application and analytical data processing. 

 

Analytics & MIS Reporting 

 

 

Data Analytics. Vehicle data is aggregated and then combined with corporate data about the vehicle and customer's policy. The data volume scales with the number of vehicles. The enriched data is stored and then sent to the usage-based insurance application to apply business-level rules to customer accounts based on results of the analytics. 

 

Dashboard Reports. SAS provides the powerful analytics engine for the usage-based insurance application and for out-of-band system analytics. Both types of analysis are important: the business process analytics to support the complex business rules of the usage-based insurance application and the system analytics to gain insights into overall system behavior. 

Below are the high-level KPIs for Dashboard Reports. 

 

  • Geolocation data, such as location of the vehicle, trip route and relevant entities along the way.  
  • Vehicle performance data, such as emissions, fuel economy, vibration, oil temperature and fuel level.  
  • Driver biometrics information, such as voice or facial recognition.  
  • Driver behavior data, such as miles/hours driven, aggressive acceleration, hard braking or erratic steering.  
  • Subscriber and registration information governing access to certain user accounts. 
  • External vehicle data, such as detecting lane boundaries, road conditions or obstacles.  

 

Vehicle Owners  

 

Real-Time Notification. Improves vehicle product quality and help OEMs securely transport critical data for over the air (OTA) updates. Reduces OEM service cost and saves time for vehicle updates, while providing a better driver experience. 

Other Mobility Services. Like Vehicle Health Management, Repair Opportunities, Events & Promotional Notifications Enables proactive service alerts and preventative maintenance diagnostics to help reduce warranty cost, improve product quality and build brand loyalty. Uses remote diagnostics and safety services to help prevent breakdowns and increase vehicle quality. 

 

Manufacturer Enterprise Users 

 

 

Internal Access. Connected vehicle solutions that quickly and securely identify, authenticate and connect users, devices and information with an open, developer-friendly and enterprise-class platform. 

Mobile Access. Monitoring and analyzing connected vehicle data allow vehicle manufacturers to offer new services to both consumers and the broader vehicle ecosystem, including suppliers, insurance companies and fleet management companies. 

 

Solution Benefits 

 

The analytics-driven growth opportunities described earlier are all achievable today, thanks to five key technologies:  

Streaming analytics. Filters and analyses data in motion by processing huge volumes at very high rates (in the range of millions per second) − with extremely low latency (in milliseconds), often embedded in devices at the edge of the network.  

Real-time decisioning / real-time interaction management receives streaming data about an event of interest − such as a Vehicle’s constantly changing location, direction, destination and context − and provides a recommendation engine that triggers the right offer or notification to the driver/dealer/insurer/manufacturer.  

Big data analytics. Taps into a collection of predictive models based on real-time or batch ingestion of relevant data in a distributed computing environment.  

Data management. Capabilities, such as data standardization and intelligent data filtering, take IoT data − generated anywhere − and make it clean, trusted and ready for analytics.  

MIS Reporting. Ensures consistency in how analytical models are being managed and monitored − and tracks the evolution of models and ensures that model performance does not degrade over its life cycle.  

 

Comments

Delighted to see the final outcome of the use case.. All the best team 🙂 

Hi, Did you build this on SAS or is it currently just on a different platform? your diagram is not clear if you just have the idea to build on SAS or is it already done as part of this Hackathon?

Hi Shyam,

We have leverage SAS Engine for Advanced Analytics based driver behavior basis

- Driving patterns (Rash driving etc.)

- Trip based analysis for considering Vehicle conditions

- Vehicle telematics data etc.

Front-End UI is built non-SAS. Although, data is flowing through SAS from back-end.

 

The alpha version is showcased in Hackthon, we are working on further versions released for industry usage. 

 

 

Nice project.  thanks for your submission.

Thanks @BillRIoT 

Version history
Last update:
‎10-20-2022 12:24 PM
Updated by:

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

Register now!

Article Tags