Through the century-long journey, the automotive industry has evolved from a traditional self-contained network to ubiquitous connectivity with the outside world. Let us look at some of the key milestones that have changed the way drivers and passengers interact.
Every other thing is getting connected to the internet today, approximately 6 connected devices per person, 120 new things are connecting every second, leading to a massive 20+ billion connected devices ecosystem. (Source- Statista, Gartner, Secondary research).
In the automotive space, Gartner predicted that by 2020, the average connected vehicle would generate over 280 petabytes of data, with at least 4 TB of data being generated in a day, 470 million connected vehicles will hit the roads by 2025 (source- PwC).
The biggest challenge here is to efficiently process & deliver this high-volume data with existing communication networks (even if it would have 5G available everywhere) & cloud computing ecosystem. Also, the challenge is not limited to handling ever exploding data generated by these connected vehicles but also ensuring security, reliability, scalability, and so on. This is where Edge Computing has a key role to play.
What is Edge Computing?
Edge computing is a distributed form of computing in which computing operations (like data collection, analysis, etc.) are performed on a device, where it is created, such as a robot, an industrial boiler, or a vehicle. This, in turn, reduces the need to transfer the data back and forth from the cloud.
Adoption of Edge Computing?
The global edge computing market size is expected to reach USD 28.84 billion by 2025, with a CAGR of 54%, says Grand View Research.
Increase in load on the cloud infrastructure, rise in Internet-of-Things (IoT) solution deployments, the advent of 5G, increasing need for low-latency, and lack of real-time applications are some of the critical drivers for the growth of Edge computing.
Edge computing has been penetrating industrial, energy, healthcare, agriculture, retail, data centres, and wearables. However, still today, smart homes, buildings & factories are the hosts to key primary use cases of edge computing. Field devices/sensors talking via communication protocol are a good fit for edge computing.
A typical industrial network contains thousands of field devices, sensors, actuators generating a massive amount of data, but with edge computing, the data can be processed at the field device for the analysis and decision making. Likewise, in connected lighting space, analytics run at lighting node can bring insights like energy usage, space occupancy, carbon footprint, and so on. Similar to this, edge computing has numerous use cases in automotive across various sub-systems including telematics, infotainment, ADAS, electric vehicle charging stations, and so on.
Edge Computing in Automotive
Historically, the adoption of computing (be it cloud or edge) and software in automotive has trailed the in-general adoption in other industries. However, the situation is not true for today, and something exciting is happening now.
Cloud computing has been around for a while in many industries and many forms. But, vehicle telematics became one of the top use cases adopted in automotive somewhere in 2008.
Connected vehicles will continue to evolve at an exponential rate with V2V and V2X communication. This generates a large volume of data (every connected vehicle will generate data up to 4TB/day). How to handle, process, analyse the large amounts of data and make critical decisions quickly and efficiently?
Automobile makers are focused on leveraging edge computing to address these ever-evolving challenges. A group of cross-industry global players has formed the Automotive Edge Computing Consortium (AECC) to drive best practices for the convergence between the vehicle and computing ecosystem. Let’s see how we can leverage edge computing in the automobile sector.
Some of the possible use cases of Edge Computing in Automotive
Edge computing offers a wide variety of use cases across automotive sub-systems right from the sensor, applications to the end-user experience. Let’s go deep into some of the key possible use cases of Edge computing specific to automotive.
1. Sensor Data- Less is more
In general, there are numerous sensors built-in everyday smart devices. From a proximity sensor in smartphones to smoke sensors in intelligent buildings/homes. Though these sensors differentiate themselves by type, technology, form-factor, etc. However, they all have one common element, which is “Data,” which is generated in large quantities.
The same applies to vehicles. A typical luxury vehicle contains hundreds of sensors. Although all the data generated is processed in the car, various in-car applications require the transfer of data to the cloud. With Edge computing, data pushed to cloud could be limited more smartly. It can process and analyse most of the data at the edge and only select non-sensitive data can be transferred to the cloud. This brings data transmission costs down and also protects the sensitive data leaving the vehicle.
2. Electric Vehicles
a. Battery monitoring & predictive maintenance
The battery of electric vehicles needs to deliver throughput in the best possible ways and to achieve this, continuous monitoring and predictive maintenance of battery are required. The health of battery depends on various factors like driver habits, acceleration, traffic conditions, charging cycles, and so on. Edge computing can aggregate all this data and perform a real-time inspection of key battery parameters and alert the vehicle owner in case of any deviation.
Edge computing could aggregate data and in real-time monitor the key battery parameters and sensor data. By leveraging edge computing, auto OEMs and network providers can make a direct influence on the customer experience, especially in the electric vehicle segment.
b. Charging Stations- Predict & Plan
Edge computing plays a key role in overall planning & optimization of charging processes, including wait time at the queue, fare, etc. This, in turn, helps in achieving greater efficiency for charging stations and thus, overall mobility.
3. Smart Traffic Management
Consider a real-life scenario at a traffic stop, especially when the intersection is a culmination of five-six roads which are heavily used most times. Only the vehicle (although if it’s “Autonomous”) would not control the long waiting times as it requires to follow the traffic norms. But, let’s consider a futuristic scenario if the road intersection has an edge device deployed to which all vehicles can communicate while coming towards the intersection. The edge device can aggregate the data from nearby vehicles and also notify them well in advance about the situation at the intersection. Thus, edge computing increases efficiency and throughput at the complex road intersections.
4. Personalized Infotainment and/or Reconfigurable Cluster
Over the last decade, In-vehicle infotainment and cluster systems have transformed the user experience in many ways through swanky HMIs, augmented reality, mobility apps, personalization and so on.
Edge computing can take the user experience to the next level by understanding applications and interfaces the user is using and where the interaction design should be optimized whether it is touch interface or voice recognition and so on.
Machine learning algorithms can continuously gather, process and gain insights from the available data. These Machine learning models can be hosted on the Edge device to analyse the sensor & user behaviour data and use this processed data to improve in-vehicle user experience.
DOWNLOAD CASE STUDY
End-to-end Product engineering for a Telematics Device
4. Vehicle Security- Multi-level authentication
Multi-level authentication could a by having multiple sensors like cameras, proximity sensors. The aggregated data from these edge devices could be used to enable multi-level authentication in which the camera will be used for face recognition, Bluetooth sensor to detect the proximity of driver’s phone, and an infrared camera for spoofing detection.
6. Fleet Management
To give one example of a fleet management company operating in commercial vehicle space, the fleet manager needs to keep all the data related to vehicle health, servicing/maintenance due before dispatching the vehicle for a trip. What if he/she sends the vehicle which is on service due after 50 miles and booked trip very likely to travel 2000 miles. Edge computing can predict such future situations by processing sensor values and analysing the machine learning models inside the vehicle. The machine learning models can be trained/refined in the cloud and pushed to Edge with over-the-air updates.
7. Predictive maintenance
Powered by predictive analytics models, Edge computing can continuously monitor various parameters of the vehicle, such as ambient temperature, mileage, tire inflation, braking, acceleration, and speed/force. The analytics model will predict if any component/part will likely to fail and alert the vehicle owner. For example- Tire pressure/condition below the safe level. Edge device will alert/remind the vehicle owner for maintenance or replacement of the tire.
Benefits of Edge Computing
- Achieve higher processing speed– Processing data closer to the source reduces network latency thus increases network performance and speed for end-user
- Increased Security– More data processed at the local device, therefore, which reduces security attacks which happen during data transfer over the network. Also, as it distributes processing, storage, and applications across a range of devices to lower the security risks significantly
- Cost savings– As edge computing retains most of the data at the device itself, it reduces the network latency which directly translates into dollars
- Superior reliability– Local storage and process ensures continuous operations & does not impact just because of most common issues like lost connectivity to the cloud
- Scalability– By bundling computing, storage, and analytics capabilities into devices, edge computing enables companies to scale-up their solutions reach and skills quickly and efficiently
Possibilities for the adoption of edge computing are enormous in automotive and across vehicle sub-systems. With the rise in connected & autonomous vehicle, the processing and analysis of the vast amount of data will be crucial in taking critical decisions which will make the vehicle safer and efficient. Edge computing will play a significant role in achieving all this in the coming years.
eInfochips leverages its expertise in product engineering, IoT, edge & cloud computing, AI/machine learning to deliver edge-based solutions across ADAS, telematics, fleet management, smart parking & traffic management segments. To know more about eInfochips, connect with us.