4. Emerging Technologies in Automotive
Industry
Edge Analytics
Fifth Generation
Wireless Technology
Edge Artificial
Intelligence
Autonomous Driving
Level 5
5. Organizational and Resource Issues
Organizational Issues
● Emerging technologies, changing customer demand and supplier issues are
some of the concerns.
● The main challenge is the integration of the various analytical systems,
hardware devices, information systems, and wireless connectivity.
Resource Issues
● The resources required for the adoption of the four emerging technologies in
automobiles are computationally intensive and expensive.
● The sourcing, cost, and testing of the hardware devices and networking
equipment that are required for implementing machine learning, and fifth
generation technology is a big challenge.
6. Risk and Change Management Issues
The risks related to the adoption of emerging technologies in automobiles are as
follows:
● Attackers can hack into vehicular data and driver data either directly or remotely
● The user’s sensitive data is at the risk of theft or misuse by hackers.
● The safety of the passengers is at risk when the hardware devices or
autonomous driving function fails.
Change management issues are as follows:
● A change could be anything from a hardware replacement, design adjustment,
installation of software to a slight change in the vehicle.
● Every change should be evaluated by the experts and engineers to make sure
that the impact of the changes should not affect the performance and safety
regulations of the vehicle.
8. Advantages and Use Cases
Advantages of Edge Analytics in Automobile Industry:
● Stronger Hardware
● Scalability
● Reliability
● Privacy
● Better Latency
Use Cases or Applications of Edge Analytics in Automobile Industry:
● Fleet Management:
● Connected Cars
● Automotive Maintenance System
● In-vehicle Infotainment and Telematics
9. Implementation
Challenges and
Recommendations for
Edge Analytics in
Automotive Industry
Implementation Challenges
● Technology
● Regulations and Standards
● Setting up of Ecosystem
Recommendations
● Overcome technological limitations
● Data gathering, storage, and
segregation
● Collaboration with Cloud
17. Advantages and Use Cases
Advantages of Edge Artificial Intelligence in Automobile Industry:
● Split-Second Decision making through advanced data processing
● Low Power Consumption
● Lower dependence on interconnectivity through cloud
● Adaptive learning for better driving experience
● Better Latency
Use Cases or Applications of Edge Analytics in Automobile Industry:
● Autonomous Driving
● Advanced Driving Assistance
● Auto maintenance of Vehicles
18. Implementation
Challenges and
Recommendations for
Edge Artificial
Intelligence in
Automotive Industry
Implementation Challenges
● Nascent Industry & Ecosystem
● Scarcity of Technical Expertise in
Artificial Intelligence
● Lack of economies of scale
Recommendations
● Develop solid technical team
● Create learning models to validate
concept
● Collaboration with other Edge
devices
20. Alignment of Business Strategy with
Technology Initiatives
● The business strategy of automotive
companies is to achieve sustainability
and customer satisfaction through
product differentiation and
technological innovation.
● Many companies aim to disrupt the
automotive sector by improving
product design, increase safety and
security measures for users.
● Hybrid Cloud Computing
Models
● Setup a software-hardware
Infrastructure
● Data collection, storage and
segregation
● Implement Fronthaul and
Backhaul Architecture
21. Implementation Timeline
January 20XX
Lorem ipsum dolor sit
amet, consectetur
March 20XX
Lorem ipsum dolor sit
amet, consectetur
June 20XX
Lorem ipsum dolor sit
amet, consectetur
July 20XX
Lorem ipsum dolor sit
amet, consectetur
October 20XX
Lorem ipsum dolor sit
amet, consectetur
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
22. Summary Justification
● Emerging technologies will help grow the edge computing
technology and improve networking capabilities.
● Technologies are dependent on one another, so an autonomous
ecosystem enables this interaction to be smooth and effective
● Fifth-generation wireless technology will help with networking
capabilities.
● Edge Analytics will enable faster communication using the sensors
attached to the self-driving vehicles
● Edge algorithm will improve the computing power
● Autonomous driving level five will emerge as a fully automated car
that will need no human intervention
Akanksha: Automotive Industry consists of all the companies and organizations that are involved in the manufacturing of motor vehicles. It plays a significant role in deciding the economy of a country. In fact, by revenue, this industry has the highest earning economy.
Akanksha: Emerging technologies like edge analytics, fifth-generation wireless technology, autonomous driving level five and edge algorithms will play a significant role in shaping the future of the automotive industry.
Edge Analytics is a way of data analysis to enhance speed. The analysis is done at the network edge like near the sensor, switch or any other connected device. This enables the data analysis at the site itself which is then sent to a centralized system for further processing.
The fifth-generation technology will have a major role to play in the network connections for self-driving cars. We know that autonomous vehicles are reaching the mass market. This means that the network should have the bandwidth to support the connectivity of hundreds of people simultaneously.
Edge Algorithms are deep learning algorithms that are highly complex models and are considered to be part of machine learning and artificial intelligence. It has the ability to run on a system, locally and send the data in real-time.
Autonomous driving level five promises full driving automation that requires no human intervention. Components like steering wheels and pedals will be eliminated at this stage. The car at this level is able to do any task that a human driver can. It can do anything and can be directed to any place.
Sai Subhasree Pakina - The automotive industry is a very large industry and for any change to occur, it has to be replicated for a large number of users. Moreover, this is not something that can be changed or improved in patches, the vehicle needs a great amount of modification for which replacement might be the easiest option. Apart from this, the cost is also way too high as with the existing technology which will again lead to hindrance in the adoption of it.
The resources that are required for implementing all the four technologies in the automotive industry are related to hardware and network requirements. The cost of high-quality edge devices with a high computational power and storage capacity would be high. The data gathered by the sensors from the vehicles and from traffic management stations is also considered as an important resource. The security of this resource is the main concern. These resources add up to the cost of implementation and hence the industry needs to invest a huge amount to adopt the aforementioned four technologies. Customer demand may not be able to offset the cost of resources.
Sai Subhasree Pakina - There are many potential issues and cybersecurity threats related to the aforementioned four technologies in connected cars. Vulnerabilities may exist in a car’s communication function, a mobile device or any other hardware device connected through a network. These vulnerabilities pose a critical threat to automobiles.
A lot of changes occur in the life cycle of an automobile. The changes made in the automobile manufacturing process should be recorded properly. The implications of these issues are severe. It could even lead to loss of life in the case of Autonomous Driving Level 5 driving where the vehicle is fully autonomous without any human intervention.
Sai Subhasree Pakina - The automobile industry is one of the fastest growing industries for the application of the Internet of Things (IoT) solution. In Edge Analytics, data is processed and stored in the sensor or device location. It works on real-time data that is gathered by the sensors or devices worn by users. Edge computing works on instant data and there is low latency.
The adoption of Edge Analytics in the automotive industry has opened up new doors for automotive companies to achieve sustainability. There are numerous applications such as traffic management, transportation, self-driving, and connected cars, etc. Connected vehicles utilize cellular networks for connectivity and communication. These vehicles have many sensors that collect real-time data to process and take the necessary action that aids in intelligent driving.
Sai Subhasree Pakina - The advantages of Edge Analytics are as follows:
Specialized Hardware: Edge devices provide stronger hardware capabilities and higher computational power for machine learning applications.
Scalability: New Edge Nodes can be added very easily. The size of hardware devices is small and it helps in the adding, updating or deleting of nodes from the network.
Reliability: Edge networks are more reliable as the data stays within the network and it is not transferred to an external server.
Privacy: The privacy of users’ sensitive data is protected by processing the data in the Edge Network itself.
Low Latency: Edge Nodes result in lower latencies as data is instantly processed and computed in real-time.
The applications of Edge Analytics are as follows:
Fleet Management: Sensory data is gathered from a fleet of vehicles to analyze and perform computations. The data is stored on the Cloud to run deep learning algorithms and create visualizations.
Connected Cars: Complex algorithms can be run at the edge nodes in the car. It also helps in traffic management when there is an edge node at junctions or road intersections
Automotive Maintenance System: Car manufacturers can utilize Edge Computing to utilize usage data for battery charging and maintenance based on sensor values and real-time data
In-vehicle Infotainment and Telematics: Users can integrate their vehicles with smartphones and other devices to enable smart warning systems, emergency braking, theft alarms, etc. Telematics can even call officials in the case of theft or forced entry in the vehicle.
Sai Subhasree Pakina - The implementation challenges are as follows:
Technology: The critical challenge is to perform machine learning computations accurately in the edge environment. So, it is challenging to perform deep learning algorithms that require a high amount of processing in the sensors at the edge.
Regulations and Standards: Autonomous cars with intelligent motion sensors need to ensure all safety measures. The challenge here is to establish safety standards and regulations for the manufacturers so that they can strictly adhere to the guidelines.
Setting up of Ecosystem: One of the challenges for automobile companies is to partner with the Internet of Things (IoT) hardware manufacturers to develop the edge network.
The recommendations for the automotive industry are as follows:
Overcome technological limitations:It is of utmost importance for the automobile companies to first tackle technological limitations such as improving the computational capacity of sensors, invest in high-quality sensors, develop algorithms for edge and test them multiple times.
Data gathering, storage, and segregation: Investments should be made in gathering historical vehicular data by automobile manufacturers. Previous data will lead to trends and patterns which will help in devising Edge Analytics algorithms to be more efficient. The data should be stored in large repositories and it should be used for training and testing of algorithms.
Collaboration with Cloud: Automotive companies should consider investing in a cloud-based platform to store all the data gathered by their vehicles, that is required for the initial testing and training. The technology infrastructure can be easily scalable, and it would allow for detailed visualizations.
When such a technology is implemented, we can improve safety as the system will be driving the car. Hence, there will be less stress among drivers. We can also design our system to work on maximum fuel efficiency. Also, the traffic can be managed in a better way as there will not be any breakage of traffic rules and hence lesser congestion. It will further lead to lower accidents, and cars will be environment friendly. Autonomous driving will also promote car sharing, which can further reduce emissions.
Since the technology is not yet mature, the production cost will be high as there will need to be intensive research which can be tackled through collaboration among companies. Also, if at all, the system fails, the risk is very high, and hence we need to have failsafe options. Also, we need to ensure that the data is not misused, and thus, we need high-security standards that can make sure that the data is safe.
Akanksha: What is fifth generation wireless technology?
According to a set of international standards, 5G or the fifth generation of mobile communication is designed to be a hundred times faster and more reliable than the fourth-generation Long Term Evolution (LTE) technology. This will eliminate any latency during data transmission.
Akanksha: This technology will have various advantages and impact on the automotive industry. The major advantages include:
Ubiquitous connectivity - that can create and better-automated cars. With our new technology, the automotive industry is trying to make V2X (Vehicle to Everything) communication that could support reduced latency of up to one millisecond.
Automated Driving: This requires a high level of accuracy which can only be achieved with reliable and accurate transmissions at very high bit rates.
Traffic Efficiency Services: The high network can enable the vehicle to take a real-time action based on the fast data transmission signals.
Road Safety: The vehicle to vehicle transmission is one of the most beneficial communications in this scenario. This detects a vulnerable road user that includes pedestrians, cyclists, and anyone with a mobile device.
Some impacts are:
Better navigation: It will be possible to use real-time video feedback and enhance the experience by augmented reality.
Automotive ecosystem: A new ecosystem will emerge due to a need to invest highly in analytics, cloud-based systems, and machine learning technology.
Digitization of Logistics: The computation of the vehicles will need communication with remote servers along with using local units to enhance the processing power. Fifth-generation networking would help to do that.
Akanksha: There are many implementation challenges for this technology. Some of which are:
The initial cost of investment for this technology is very high and which partner should bear this cost creates major hurdles in its development.
The devices do not support the right antenna positioning to find the wavelength gain and loss that is essential for transferring the fifth-generation networking signals.
There are strict regulations on the amount of personal data that can be shared. With huge data transmission, there might be security-related issues. Thus, regulators will impose strict rules on the network operators, creating further implementation challenges.
Few of the recommendations are:
Create in production robots that can be used for body car construction saving a lot of manual effort and labor cost. It will also improve the overall telematics’ system.
Work to enable vehicle interaction that is important to reduce accidents and will push down the maintenance and insurance costs.
In-vehicle infotainment has huge potential as customers will be interested in services like in-vehicle retail, vehicle-to-vehicle gaming and personalized voice assistant.
Sushant Patil:
Just like Edge Analytics, the concept of Edge AI deals with computing done on one end/device. In this case, we are running artificial intelligence algorithms at edges/devices to transfer computing power to the device from cloud to have more latency and better real-time processing of data.
Sushant Patil:Edge Artificial Intelligence algorithms usually have best in class real-time data processing capabilities. This comes in handy when devices have to make split-second decisions. As such, autonomous driving will benefit from Edge AI.With Edge AI, retrieval of data, processing of data, and running of AI algorithms take place at a device instead of a cloud. Consequently, this results in lower dependence of devices on cloud. This, in turn, results in low power consumption.
By leveraging artificial intelligence and machine learning, Edge AI can provide an exquisite driving experience in an autonomous car.
As Edge AI devices process data in-house, latency is lower and devices respond quickly to change in inputs.The use-cases for Edge Analytics would be autonomous driving, advanced driving assistance, and auto maintenance of vehicles.
Sushant Patil:Since the autonomous vehicle and Edge AI industries are very young, there are no established players, suppliers, and distributors.It is difficult to build team with enough artificial intelligence expertise as the market has not matured as of now. Since the market is nascent, economies of scale won’t be achieved anytime soon.
Recommendations:
Firms should look forward to developing strong in-house AI programs.
Firms should collect historical data and create learning models to support autonomous driving features before they go into production.
Firms should look to collaborate with other organizations to tap into all the knowledge available on autonomous driving and Artificial Intelligence for edge computing.
Sai Subhasree Pakina - The following technological initiatives align and satisfy the business strategy of automobile companies:
Hybrid Cloud Computing Models: Companies should work together to build a system that is compute-intensive, storage-intensive or both. This framework is an essential part of the infrastructure wherein collaborative innovation can take place.
Setup a software-hardware Infrastructure: Partners from different industries like network, automobile, aviation, and information technology, should work together to frame a common infrastructure. This can be used for the research and development of technologies in a collaborative environment.
Data collection, storage, and segregation: It is important for an automobile company to aggressively invest in research and development. The future of robotics and self-driving cars depends on the amount of real user data gathered from the sensors.
Implement Fronthaul and Backhaul Architecture: Developments to set up a Backhaul are in place that promise to cater to complex network traffic. Efforts should be made to make it more scalable and cost-effective. For a Fronthaul, public radio interfaces are established. This should be used by the backhaul to establish network connectivity between several devices.
Akanksha: The automotive industry has been constantly evolving with the emergence of new technologies. This decade has witnessed advancements in electric cars and self-driving vehicles. To make cars become smarter and fully automated, it is important for different technologies to integrate well with each other.
Edge computing is not efficient without the usage of fifth-generation wireless communication. Autonomous driving level five can be achieved only if the vehicles can respond and analyze signals in real-time. For this, edge analytics and edge algorithms play a major role. Hence, the entire system needs to be tightly integrated and fully functional to make the automotive industry create machines that are completely free from human intervention.
Emerging technologies like edge analytics, fifth-generation wireless technology,autonomous driving level five and edge algorithms will play a significant role in shaping the future of the automotive industry. Though each technology has a huge potential in itself but
is highly dependent on each other. Edge Analytics is a way of data analysis to enhance speed. The analysis is done at the network edge like near the sensor, switch or any other connected device. This enables the data analysis at the site itself which is then sent to a centralized system for further processing. The fifth-generation technology will have a major role to play in the network connections for self-driving cars. We know that autonomous vehicles are reaching the mass market. This means that the network should have the bandwidth to support the connectivity of hundreds of people simultaneously. Edge Algorithms are deep learning algorithms that are highly complex models and are considered to be part of machine learning and artificial intelligence. It has the ability to run on a system, locally and send the data in real-time. Autonomous driving level five promises full driving automation that requires no human intervention. Components like steering wheels and pedals will be eliminated at this stage. The car at this level is able to do any task that a human driver can. It can do anything and can be directed to any place.