Robot autonomy is driving an ever-increasing demand for computational power, including on-board multi-core CPUs and accelerators such as GPUs, to enable fast perception, planning, control, and more. Careful scheduling of these computational tasks on the CPU cores and GPUs is important to prevent locking up the finite computational capacity in ways that hinder other critical workloads; delays in computing time-critical tasks like obstacle detection and control can have huge negative consequences for autonomous robots, potentially resulting in damage, substantial financial loss, or even loss of life. In this paper, we leverage recent advances from real-time systems research. We apply TimeWall, a component-based real-time framework, to the computational components of an autonomous drone and experimentally show that the timeliness and safe operation properties of a drone are preserved even in the presence of increasing interfering computational processes.

Fourier transforms are vital for a broad range of signal-processing applications. Accelerating FFTs with GPUs offers an orders-of-magnitude improvement vs. CPU-only FFT computation. However, two problems arise when executing FFT tasks with other GPU work. First, concurrent GPU use introduces unpredictability in the form of lengthy response times. Second, it is unclear how to best parameterize and schedule FFT tasks to meet the throughput and timeliness constraints of real-time signal processing. This work investigates how FFT and other GPU-using tasks can concurrently access a GPU while maintaining bounded response-time guarantees without sacrificing throughput. In our experiments, the techniques proposed by this work result in an up to 17% improvement in worst-case FFT response times.

The increasing complexity and parallelization of hardware and applications in embedded systems have brought unavoidable uncertainty in determining the worst-case execution times (WCETs) of real-time tasks. While budget enforcement can address this uncertainty by limiting an overrunning task from affecting the rest of the system, less explored is how to determine the rate and pattern of failures resulting from budget overruns. For analysis using probabilistic techniques, one must first consider potential dependence relations across different tasks or jobs of the same task. As a result, prior work on probabilistic WCET (pWCET) distributions seeking to enable independence assumptions has suffered from excessive pessimism and intricate derivation processes. In contrast, industry designs have opted for relatively simple heuristics such as “fudge factors,” budgets set by scaling mean or observed worst-case execution times by a constant factor. However, such heuristics do not have a strong analytical foundation. This paper addresses this gap in theory and practice, presenting analysis of a budgeted real-time system’s failure rate not reliant on extensive knowledge of a task’s execution behavior or independence assumptions, only requiring approximations of the mean execution time and standard deviation. This analysis bounds the rate of deadline failures, particularly those which would violate weakly-hard robustness specifications, to efficiently and optimally allocate budget and to evaluate industry heuristics.

The fixed preemption point (FPP) model has been studied as an alternative to fully preemptive and non-preemptive models, as restricting preemptions to specific, predictable locations within a task's execution can simplify overhead analysis without disallowing preemptions entirely. Prior work has produced response-time analyses for global Earliest Deadline First (G-EDF) scheduling under the FPP model. However, scheduling decisions based solely on task deadlines may be too coarse-grained and may not lead to the lowest response times. In this paper, we propose global FPP EDF-like (G-FPP-EL) scheduling, which assigns a priority point in time for each non-preemptive region of a task. We adapt compliant-vector analysis (CVA) to our model and present general response-time bounds for G-FPP-EL schedulers. We then demonstrate that it is possible to design G-FPP-EL schedulers acheiving response-time bounds optimal under CVA and argue that such schedulers should replace global FPP EDF.