SHARP: Supercomputing for High-Speed Avoidance and Reactive Planning
Kieran Lachmansingh, José Ramón González, Jacob Chisholm, Ryan Eric Grant, Matthew Pan
AI summary
Problem
Modern robots require millisecond-scale reactivity for dynamic human-shared environments, but onboard processors are constrained by size, power, and cost. It remains unclear whether HPC's massive parallelism can overcome network latency and jitter to enable real-time robotic control.
Approach
SHARP offloads collision-avoidance planning from a 7-DOF robot to local and remote HPC clusters using a hash-distributed multi-goal A* search implemented via MPI, evaluated through a high-speed projectile-dodging stress test.
Key results
- Mean planning latencies of 22.9 ms (local) and 30.0 ms (remote, 300 km away)
- Avoidance success rates of 84% and 88% against high-speed projectiles
- HPC computation ceases to be the bottleneck when round-trip latency stays under 30 ms
- A reproducible end-to-end timing and success-rate evaluation template for HPC-driven robotics
Why it matters
It validates hybrid control architectures for scalable, reactive robots, showing that low-level safety reflexes can coexist with HPC-offloaded planning in dynamic human-robot workspaces.
Abstract
This paper presents SHARP (Supercomputing for High-speed Avoidance and Reactive Planning), a proof-of- concept study demonstrating how high-performance computing (HPC) can enable millisecond-scale responsiveness in robotic control. While modern robots face increasing demands for reactivity in human–robot shared workspaces, onboard pro- cessors are constrained by size, power, and cost. Offloading to HPC offers massive parallelism for trajectory planning, but its feasibility for real-time robotics remains uncertain due to network latency and jitter. We evaluate SHARP in a stress- test scenario where a 7-DOF manipulator must dodge high- speed foam projectiles. Using a hash-distributed multi-goal A* search implemented with MPI on both local and remote HPC clusters, the system achieves mean planning latencies of 22.9 ms (local) and 30.0 ms (remote, 300 km away), with avoidance success rates of 84% and 88%, respectively. These results show that when round-trip latency remains within the tens-of-milliseconds regime, HPC-side computation is no longer the bottleneck, enabling avoidance well below human reaction times. The SHARP results motivate hybrid control architectures: low-level reflexes remain onboard for safety, while bursty, high-throughput planning tasks are offloaded to HPC for scalability. By reporting per-stage timing and success rates, this study provides a reproducible template for assessing the real-time feasibility of HPC-driven robotics. Collectively, SHARP reframes HPC offloading as a viable pathway toward dependable, reactive robots in dynamic environments.