Running Drive Cycle on the Hardware in the Loop
September 8, 2021: Amazon Elasticsearch Service has been renamed to Amazon OpenSearch Service. Run across details.
Automotive customers use AWS equally their platform for advanced driving assist systems (ADAS) and autonomous driving (AD) development to accelerate their development cycles and experience faster fourth dimension-to-market. In the blog post, Autonomous Vehicle and ADAS development on AWS Part one: Achieving Scale, we illustrated how software in the loop (SiL) and hardware in the loop (HiL) simulations are office of the workflow used to develop and validate condom AD and ADAS functionality. In this mail, I run through some of the more common questions and patterns for implementing HiL on AWS, while looking at some of the differences from running your evolution all on-premises.
HiL simulations leverage test bulldoze data and derived synthetic data to develop and validate various functions in the AD software stack. Exam drive log data is ingested and stored in Amazon S3 for use for HiL simulations in parallel with other Advert development workloads including visualization, processing, labeling, analysis, and model and algorithm development.
Every bit such, we run into customers exist in a hybrid context with their HiL workloads running on-premises to support customized equipment. For ADAS and ADS customers, this poses a few questions and considerations:
- What are the recommendations to deploy HiL for Advertisement on AWS?
- How is this different from what customers were used to on-bounds?
For hybrid customers, there are assumptions and misconceptions:
- Practice I demand to replicate the test drive data locally, and if and then, what are the considerations and consequences?
For the purposes of brevity, the remainder of this blog post volition use the term Advertizement evolution to embrace ADAS and Advertizement unless specifically called out.
HiL Building Blocks
Simulations and validations make upward an important attribute of AD development. According to Rand'south analysis, at that place is a need to demonstrate safe driving on billions of miles for an autonomous vehicle to have a lower failure rate than a human driver. While this analysis is statistically derived for fully autonomous driving (SAE Level 5), it demonstrates a need for further validation on millions of miles. This pattern is reflected in well-nigh ADAS and AD development projects, where software in the loop (SiL) and HiL are used for verification and validation.
The following diagram is an illustration of the ISO 26262 V-Model, a production development approach for matching requirements with respective tests, where HiL simulation is required for much of organisation level testing and validation phases on the right side of the overlaid V-Model.
Figure 1: V-Model equally defined by ISO-26262
HiL simulations crave a few key elements. The main component is the device under test (DUT), such equally one or more than electronic control units (ECUs) running the AD software stack. HiL simulations permit customers to put the device under test (DUT) under the rigor of real-world signals institute in a vehicle. By providing more authentic environments and scenarios for the DUT, it can be fine-tuned for primal operation indicators (KPIs) such equally ability utilization, response time, and accuracy.
The DUT is continued to a "HIL Rig," a high performance server with multiple expansion boards to connect to diverse components of the AD system. The diverse interfaces are identified in Figure 2: High Level Hardware-in-the-loop (HiL) System and Interfaces and include Controller Area Network (Can), Automotive Ethernet, Low Voltage Differential Signal (LVDS), and PCIExpress. These interfaces emulate the vehicular topology for testing purposes and allow system level validation of the DUT.
The HiL Rigs and the corresponding software tooling are offered past companies like Elektrobit, dSpace, National Instruments, and Opal-RT. The HiL systems facilitate time synchronized inputs and outputs to the DUT and measures system performance. These solutions have optimizations for large-calibration operations aligned with faster validation cycles for customers. This latter point is relevant when a projection requires validation across a large number of miles. Elektrobit provides the ability to orchestrate and deploy large HiL server farms that can be deployed to work in parallel. The larger server farms allow parallel HiL simulations to reduce the time to validate thousands of miles of drive time and assess the primal operation indicators (KPIs) for the feature sets. Results can be acted on more than quickly and reduce the overall development time.
Figure 2: High Level Hardware-in-the-loop (HiL) Organization and Interfaces
The HIL organisation loads sensor information derived from examination bulldoze logs. These logs vary in format, but frequently are captured and stored equally MDF4, ADTF, rosbag, or other information logger proprietary formats. These are then processed for HIL simulations to implement open-loop and closed-loop simulations.
- Open loop simulations refer to replay of log data from exam drives.
- Closed-loop simulations rely on the behavior of the organisation as inputs vary based on new outputs of the simulation.
Both open loop and closed loop simulations are role of autonomous driving development, but open up-loop simulations require the largest datasets (multiple petabytes on average) due to reliance on the log information from test drives making them a principal concern for deploying HiL in a hybrid manner.
Overview of Solution
Architectures for supporting HIL simulations with AWS for AD development vary based primarily on the networking bachelor at HiL locations. A mutual pattern for AWS customers is to have the HiL systems directly interfacing with Amazon S3 over loftier-bandwidth network links leveraging AWS Directly Connect. This is the simplest approach to deploying HiL and avoids hybrid data management of the petabytes of information in Amazon S3 to a local storage arrangement.
AWS Direct Connect provides customers options to deploy their HIL rigs at their information center or in AWS colocation facilities with low latency connections. AWS has the largest number of Direct Connection locations and points-of-presence (POPs) to enable depression latency connectivity to whatever of the >24 AWS Regions. The following diagram illustrates a reference architecture leveraging a direct interface from the HiL systems and Amazon S3.
Effigy three : Reference Architecture for Hardware-in-the-Loop (HiL) Direct to Amazon S3
Equally shown in Figure three, we illustrate the common interfaces, topology, and AWS services used for democratic driving customers.
- Amazon S3 is used to store and analyze the exam drive logs used past the HiL simulations and as well the results from the simulation runs for further assay.
- Metadata of test drive data is populated in various database and analytics services, referred to as the data catalogues, with metadata crawlers and processing pipelines that extract from the drive log and test result information on Amazon S3.
- The data catalogues provide flexible search interfaces for developers and validation engineers or avant-garde analytics tools. These systems provide keyword search in Amazon Elasticsearch or SQL queries in Amazon Redshift or Amazon RDS and noSQL interfaces using Amazon DynamoDB. Amazon Partner Network solutions for these database and analysis tools are common as well, such every bit those in AWS Marketplace.
- Validation engineer, data scientists, or developers apply these data catalogues to find scenarios for testing. These personas also use the HiL management interfaces to configure and orchestrate the HiL simulation runs on the scenarios identified and ensure traceability.
- HiL direction systems control the HiL Rigs that interface to the DUT and implement the HiL simulations using the examination bulldoze logs. The HiL management system then writes results back to S3 for farther assay via diverse tool bondage.
A common question AWS customers have is how to decide an optimal hybrid architecture using this arroyo. The main factors are properly sized network links to accommodate information sets used by the HiL simulations equally well as low latency network links betwixt Amazon S3 and the HiL rigs. Equally a consequence, a key factor is ensuring use of an AWS region for your AWS storage that is in close proximity to your HiL testing site(southward).
Based on current HiL implementations, open-loop simulations can sustain latencies of 30-fifty ms RTT. AWS has numerous AWS Direct Connect locations in co-location facilities with latencies <5ms RTT. Sizing for these network links tin can exist calculated based on the expected dataset sizes and the interval of time targeted for simulation run. Nosotros evidence a basic formula used for network sizing.
Average_Throughput (Gbps) = Average_Dataset_Size(GB)*8 / Time_Interval (seconds)
As an example, for a scenario where an average of 20PB is needed by the HIL rig every 2 weeks, we require ~200Gbps for the AWS Direct Connect bandwidth.
Figure 4 shows an instance of a high-level architecture supported by Elektrobit with multiple EB 9101 test racks grouped together. This architecture supports multiple ECUs to be tested at once, leveraging drive log data in Amazon S3. This system is controlled with a primal management software that allows optimal orchestration to keep the Elektrobit HiL system running optimally.
Use cases include:
- The automated replay of all relevant sensor data with high time precision to ECU
- Capture of ECU responses including debug data
- Integration of customer components inside the HiL rack for visualization or postal service-processing.
Some other mutual question from AWS customers is whether this architecture is supported for their HiL implementation. Many HiL providers are adding AWS functionality to their software and hardware stacks in response to customers transitioning to cloud for the development platforms. Some vendors still require Amazon S3 as a supported interface in their HiL Rigs. The work needed to accommodate Amazon S3 is usually a modest level of effort for any developer by using Amazon SDKs on the HiL rig software stack. If in that location is a project where this is needed, contact the AWS account teams and your HiL vendors to ensure a successful and cost efficient project implementation.
Effigy 4: Elektrobit HiL Architecture with AWS
An alternative HiL solution shown in Figure five includes Amazon S3 as the principal storage for bulldoze log data and the scale out NAS storage system is located on-premises operating as a cache for the HIL rigs. This is mutual when the networking options at the HIL site are limited in bandwidth or latency to handle the target datasets and fourth dimension windows.
AWS customers summate the size of the cache to transfer the entire dataset over the intended time interval. Following is a elementary calculation to demonstrate this.
Cache_Size(GB) = Average_Dataset_Size(GB) - Average_Throughput (Gbps) /8 * Time_Interval (seconds)
In this instance, a customer has 40 Gbps AWS Direct Connect available and a 10PB dataset needed for HIL simulations every ii weeks. Using the preceding formula there is a need for local cache of four Lead capable of high read-rates.
Figure 5: Reference Architecture for Hardware-in-the Loop (HiL) with Local Cache to Amazon S3
In this hybrid compages there is a need to orchestrate the data movement in line with the needs of the HIL simulation data prepare. This requires third party software generally or congenital in functionality into workflow orchestration tools like Apache Airflow. At CES 2019, Dell EMC and AWS illustrated a solution for this hybrid compages documented in this short solution brief using Isilon as the scale out NAS storage organisation and DataIQ as the data motion and orchestration machinery.
Whatsoever of these architectures can be price-optimized, and AWS has programs and pricing options for Amazon Direct Connect as well as the other AWS services involved. In that location are Enterprise Agreements and Migration Acceleration Programs (MAP) in line with the holistic AD development platform needs, that reduce the costs for hybrid compages functionality needed in the HiL solutions. 1 mutual need is back up for AWS Direct Connect "flat rate" pricing option to adjust the data transfer out (DTO) needs for the HiL workload. If y'all need details on these programs for your AD development project, contact your AWS account team.
Conclusion
In this blog post, we discussed 2 mutual architectural patterns for supporting HiL simulations for ADAS and Democratic Driving development. These help customers decide on the right networking, storage, and hybrid topologies for these systems.
HiL systems straight interfacing with Amazon S3 is the near mutual design as you run across with Elektrobit HiL solutions, but for customers with limited network links the employ of a local cache is an option. Autonomous driving customers looking to increase velocity in their SAE Level 2-five development programs with HiL simulations have achieved success with AWS as the development platform using these patterns. AWS has a team dedicated to autonomous driving, so contact your AWS business relationship team to get a more prescriptive solution for your HiL or related ADAS and Advertizement development needs.
Also, bank check out the Automotive issue of the AWS Architecture Monthly Magazine.
0 Response to "Running Drive Cycle on the Hardware in the Loop"
Post a Comment