Systematic evaluation of sandboxed software deployment for real-time software on the example of a self-driving heavy vehicle
DOI: 10.1109/itsc.2016.7795942
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the challenge of adopting continuous deployment and sandboxing technologies, such as Docker, in the automotive domain, specifically for safety-critical, real-time systems like self-driving vehicles. While software-only products benefit from persistent monitoring and rapid updates via containerization, cyber-physical systems face complex constraints regarding scheduling precision and hardware limitations. Motivated by the development of a self-driving Volvo FH16 truck for the 2016 Grand Cooperative Driving Challenge, the authors aim to systematically evaluate whether sandboxed execution environments introduce significant performance overhead compared to native deployment. The study employs a multi-method research approach comprising a literature review, a controlled laboratory experiment, and a real-world validation on the autonomous truck. The literature review utilized a snowballing technique to identify relevant studies on container performance. The controlled experiment measured scheduling precision and input/output (I/O) performance of sample applications across four execution environments: native versus Docker deployment, combined with vanilla versus real-time (preempt_rt) Linux kernels. Scheduling precision was assessed using a "Pi" calculation task, while I/O performance was measured via camera capture and disk storage tasks. System load was simulated using stress-ng to mimic operational conditions. These results were validated in an uncontrolled experiment on the actual truck, focusing on scheduling precision under realistic system load. The findings indicate that the choice of deployment strategy (native vs. Docker) has a negligible impact on system performance. Statistical analysis (MANOVA) revealed that the kernel type had a significantly greater effect on scheduling precision and I/O performance than the deployment method. Specifically, using a real-time enabled kernel improved determinism and reduced latency, whereas the overhead introduced by Docker was statistically insignificant. In the controlled experiments, scheduling precision and I/O metrics showed no noticeable difference between native and sandboxed execution. The real-world validation on the truck confirmed these results, demonstrating that Docker does not compromise the real-time capabilities required for autonomous driving when paired with an appropriate kernel. The significance of this work lies in its validation of container-based technologies for cyber-physical systems. The authors conclude that recent trends in software architecting, such as microservices encapsulated in sandboxes, can be safely applied to automotive software engineering without incurring prohibitive performance costs. This enables the automotive industry to adopt continuous integration and deployment practices similar to those in cloud-based software, facilitating faster updates, better maintenance, and improved software lifecycle management for self-driving vehicles. The study highlights that while Docker is viable, the selection of a real-time kernel remains the critical factor for ensuring deterministic performance.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-25 |
| archive | success | unpaywall | — | — | 2 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | success | openalex | — | — | 1 | 2026-06-26 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: tool software