To ensure the safety of ADAS/AD systems, using recorded real-world data in the homologation and release strategies is essential. After all, contrary to synthetic simulation data, the level of fidelity of real-world data is unquestionable.
dSPACE offers a flexible solution that uses the replay of recorded data (also referred to as data replay, data reprocessing, data re-simulation, and data playback) for the validation and testing of ADAS/AD functions, including perception and sensor fusion components.
Key Benefits
- Easy extension of an existing HIL system to a data replay solution
- Synchronous sensor and bus data ingestion with minimal jitter
- Support for a wide range of standard and customer-specific bus and sensor interfaces
- Real-time adaptation of the recorded data with security and time information
- Interoperability between a pure software environment (SIL) and a physical SUT (HIL)
- Live monitoring and manipulation of the replayed data
dSPACE offers dedicated solutions for all data replay use cases, whether the SUT (system under test) is physically available (HIL) or still in the early software development phase (SIL), and whether it is a simple sensor component or a complex central computation platform for autonomous driving.
All dSPACE data replay solutions ensure synchronous sensor and bus data ingestion with minimal jitter. Moreover, the modular approach makes it easy to perform system upgrades to meet new bandwidth requirements. Seamless transition and interoperability between a pure software environment (SIL) and a physical SUT (HIL), and between closed-loop and data replay tests is also ensured. This interoperability simplifies feeding the SUT with closed-loop synthetic data to reach the initial state of the recorded data. Then the data replay process starts.
dSPACE data replay solutions support a wide range of standard and customer-specific bus and sensor interfaces. Moreover, by using the dSPACE real-time systems, you ensure that the recorded data is continuously updated with security and time information in real time, and you can perform live monitoring and manipulation of the replayed data.
Replay Accuracy
The aim of data replay is to reproduce recorded data in exactly the same way in which the (V-)ECU would have received it if it was used in a vehicle on the road. dSPACE understands this challenge very well and offers best-in-class solutions, allowing you to focus on innovations that set you apart from your competition. Simply leave the rest to dSPACE.
Modularity
Built with the fast dynamics of ADAS/AD projects in mind, each dSPACE solution consists of modular blocks that can be rapidly and efficiently adapted and expanded. The modular system architecture gives you a head start over your competition, as you can expand the system later on, for example, once the initial system is at full capacity or when a completely new feature is required.
Comprehensiveness
Validating ADAS/AD systems by means of data replay is a demanding process, not just a simple test step. After all, complex software data replay test campaigns are required for each software update, and specific hardware data replay test campaigns have to be completed for the main scenarios of interest. And we haven’t even mentioned test and data management and the corresponding IT infrastructure. dSPACE recognizes these challenges and offers the right solution for each of them, whether we tap directly into our own dSPACE expertise or work with our extensive partner ecosystem. You can choose the best fit for your company from the market's most comprehensive portfolio.
Optimized Total Cost of Ownership
dSPACE looks at the entire test chain from beginning to end to assess and optimize what you will really end up paying throughout the lifetime of a solution. dSPACE stands for full transparency. We avoid hidden costs that result from a lack of essential features that would make accurate testing infeasible, and we actively seek to decrease computation/usage costs. In addition, with our extensive know-how in closed-loop testing, all dSPACE systems for data replay can be used again as closed-loop systems with little to no effort.
Challenges
Achieving high precision and realism is a demanding task. The system under test (SUT) has to be fed with the recorded data exactly as it would be during a real test drive. Heterogeneous data streams, including sensor and network/bus data, have to be synchronized during the data replay process, despite their different package/message size. And the sheer amount of streamed data is steadily increasing. Another challenge is the complex real-time adaptation of the recorded data to pass the end-to-end encryption and security checks in the SUT. Last but not least, efficiently scaling the tests to continuously feed millions of driven kilometers into the system and to orchestrate multiple test systems is crucial.
