RoboGPU: Accelerating GPU Collision Detection for Robotics

TL;DR

The proposed RoboGPU architecture introduces a RoboCore that computes collision queries 3.1 times faster than RTA implementations and 14.8 times faster than a CUDA baseline, and matches the performance of dedicated hardware accelerators while being able to adapt to evolving motion planning algorithms and support classical algorithms.

Abstract

Autonomous robots are increasingly prevalent in our society, emerging in medical care, transportation vehicles, and home assistance. These robots rely on motion planning and collision detection to identify a sequence of movements allowing them to navigate to an end goal without colliding with the surrounding environment. While many specialized accelerators have been proposed to meet the real-time requirements of robotics planning tasks, they often lack the flexibility to adapt to the rapidly changing landscape of robotics and support future advancements. However, GPUs are well-positioned for robotics and we find that they can also tackle collision detection algorithms with enhancements to existing ray tracing accelerator (RTA) units. Unlike intersection tests in ray tracing, collision queries in robotics require control flow mechanisms to avoid unnecessary computations in each query. In this work, we explore and compare different architectural modifications to address the gaps of existing GPU RTAs. Our proposed RoboGPU architecture introduces a RoboCore that computes collision queries 3.1$\times$ faster than RTA implementations and 14.8$\times$ faster than a CUDA baseline. RoboCore is also useful for other robotics tasks, achieving 3.6$\times$ speedup on a state-of-the-art neural motion planner and 1.1$\times$ speedup on Monte Carlo Localization compared to a baseline GPU. RoboGPU matches the performance of dedicated hardware accelerators while being able to adapt to evolving motion planning algorithms and support classical algorithms.

Figures (19)

• Figure 1: SIMT efficiency and number of collision queries required by each device on four environments from M$\pi$Net mpinet (left). Required computation for the average collision query (right).
• Figure 2: Example point cloud from the Tabletop environment of M$\pi$Net mpinet, with the robot's OBBs shown in red.
• Figure 3: Separating axis collision test (SACT) between AABB from environment octree and OBB from robot. If a separating line divides the projections of the two boxes, then the boxes are not colliding. Reproduced from energyefficientmotionplanning2024shah.
• Figure 4: Baseline GPU architecture. RoboGPU replaces RTA with RoboCore.
• Figure 5: Ray tracing accelerator architecture on a GPU, which may include fix-function intersection hardware (RTA design) or programmable operation units (TTA$+$ design ha2024tta) to execute intersection tests.