CudaRobotics Docs

Evidence index

Results and Reproducibility

CudaRobotics keeps benchmark reports in the repository so wins, negative controls, and limitations are visible next to the code.

Distribution Results

Start here for the published v0.3.0 release evidence.

Nav2

CUDA MPPI vs Nav2 CPU MPPI

Closed-loop pluginlib comparison on the same costmap, plan, and limits.

Open report

Nav2

Extended controller scenarios

Double-gap and moving-crossing smoke coverage for path-window and dynamic-costmap regressions.

Open report

Critic

Path-angle critic

GPU MPPI heading-to-path tangent cost for high-curvature tracking.

Open report

Critic

Curvature speed critic

Optional path-curvature speed target for smoother bend entry.

Open report

Critic

ESDF clearance critic

GPU-only comparison of default costmap scoring and optional distance-field clearance.

Open report

Registration

External baselines

FilterReg comparison against probreg and Open3D CPU baselines.

Open report

Registration

Unified GPU smoke

Five GPU registrars under one schema with explicit quality gates and environment metadata.

Open report

Real data

Recorded rosbag evaluation

Recorded/shadow evidence with command pairing, clearance, and a retained failed quality gate.

Open report

Release

v0.3.0 release checklist

Required CPU, GPU, ROS, packaging, and evidence gates before the tag is published.

Open checklist

MPPI Zoo

The fixed-seed suite compares the baseline MPPI planner against a growing set of paper-inspired variants. It is not a universal claim; it is a reproducible local stress suite with both successes and failures checked in.

Report Scope
2026-06-10 suite Five scenarios, ten planners, multiple sample counts.
Smoke suite Small quick-run suite for fast regression checks.
Reproduction zoo Research backlog with implementation notes and limitations.
MPPI zoo fixed-seed suite chart
Checked-in MPPI zoo chart generated from the repository benchmark scripts.

Release Notes