zpe-robotics 0.1.1

Deterministic motion utility kernel for frozen wire-v1 robotics transport and replay.

ZPE-Robotics

SAL v7.0 — free below $100M annual revenue. See LICENSE.

---

What This Is

187× compression on real robot joint streams (vs zstd_l19 4.59×, vs zstd_l3 4.44×) — with PrimitiveIndex search and VLA token export. Governing benchmark: LeRobot real-data suite, 3 datasets across 3 families. Proof path: proofs/enterprise_benchmark/GATE_VERDICTS.json.

ZPE-Robotics compresses and replays joint-stream logs for infrastructure teams that need searchable motion archives instead of raw bagfiles. The 187× figure is real and bounded-lossy (smooth-trajectory slices; step inputs cause Gibbs ringing — see footnote). Engineering surface remains blocker-governed: step-sequence behavior, comparator closure, and runtime closure are not closed.

Wire encoding: wire-v1 (frozen packet contract; authority: docs/ZPBOT_V2_AUTHORITY_SURFACE.md)

Key Metrics

Metric	Value	Baseline / Notes
COMPRESSION	187׆	LeRobot real-data benchmark; full baselines in Competitive Benchmarks section
ENCODE_P50	0.11 ms	per 1 000 frames, gate B4 PASS
DECODE_P50	0.089 ms	per 1 000 frames, gate B5 PASS
VLA_TOKEN_EXPORT	24-token FAST surface	vla_bridge.py
PRIMITIVE_SEARCH P@10	1.0	on REACH template, synthetic corpus
ANOMALY_FPR	0.05	recall=0.9 at threshold=3.22; Phase 10 holdout (100 nominal / 10 anomalous)
CROSS_PLATFORM_PARITY	PASS (3 lanes)	arm64-qemu / macOS / ubuntu-x86 produce identical SHA256
BENCHMARK_GATES	4/5	3 datasets, 3 families; B3 (bit-exact) fails

Sources: proofs/enterprise_benchmark/GATE_VERDICTS.json (B2: 40.74× vs zstd_l19, B4/B5 latency) | proofs/red_team/red_team_report.json (42.14× vs zstd_l3, attack-5 FPR) | proofs/release_candidate/primitive_search_result.json (P@10) | proofs/release_candidate/anomaly_detection_result.json (FPR/recall) | proofs/release_candidate/it04_parity_matrix_result.json (cross-platform)

† Bounded-lossy. The ≤ 0.5° angular figure is limited to smooth-trajectory slices; it is not a general motion bound. Step/discontinuous inputs cause Gibbs ringing, with 68° RMSE measured on a unit-amplitude step signal. Baselines are lossless.

Competitive Benchmarks

Competitive benchmark evidence: proofs/enterprise_benchmark/benchmark_result.json | proofs/red_team/red_team_report.json | proofs/artifacts/lerobot_expanded_benchmarks/aggregate_spread_summary.json

Tool	Compression Ratio	Notes
ZPE P8	187.13׆	governing LeRobot real-data benchmark; PrimitiveIndex search requires decode
zstd_l19	4.59×	strongest retained classical codec; ZPE is 40.74× better (gate B2 PASS)
zstd_l3	4.44×	red-team attack 1 baseline; ZPE is 42.14× better
gzip_l9	3.97×	retained gzip baseline
mcap_zstd	3.99×	MCAP container baseline
lz4_default	3.00×	low-latency baseline
h5py_gzip9	2.69×	HDF5 gzip baseline
h5py_lzf	2.15×	HDF5 fast baseline

† Bounded-lossy. The ≤ 0.5° angular figure is limited to smooth-trajectory slices; it is not a general motion bound. Step/discontinuous inputs cause Gibbs ringing, with 68° RMSE measured on a unit-amplitude step signal. All other baselines are lossless.

What We Prove

Auditable guarantees backed by committed proof artifacts. Start at docs/AUDITOR_PLAYBOOK.md.

• Spectral wire transport with directional reasoning layer for robot action sequences
• Search operates on decoded motion streams via PrimitiveIndex
• Red-team resilience: 4 attacks withstand, 1 fails, 1 partially withstands, 1 remains open
• VLA tokenization aligns with vision-language-action model input formats
• Public package acquisition route is pip install zpe-robotics; package availability does not change blocker status

What We Don't Claim

• Full release readiness
• Bit-exact .zpbot round-trip replay
• B3 benchmark gate pass
• Red-team attack 3 lossless qualification
• General anomaly readiness beyond the declared threshold-selected holdout surface
• Robotics Rust ABI
• Independent third-party reproduction
• Generally valid ≤ 0.5° angular fidelity — the figure comes from smooth-trajectory slices only; FFT-based encoding causes Gibbs ringing on step/discontinuous inputs (68° RMSE measured on a unit-amplitude step signal)
• Search-without-decode — PrimitiveIndex requires full packet decode before indexing

Commercial Readiness

Field	Value
Verdict	BLOCKED
Authority	proofs/ENGINEERING_BLOCKERS.md

Selected Claim Gate

Field	Value
Gate	bounded-lossy archive/search
Gate Status	PASS for the narrow claim only
Full Engineering	BLOCKED
Proof	proofs/narrow_claim/NARROW_CLAIM_GATE.json

Tests and Verification

Code	Check	Verdict
V_01	BENCHMARK_GATE_B1_COMPRESSION	PASS
V_02	BENCHMARK_GATE_B2_ZSTD_BASELINE	PASS
V_03	BENCHMARK_GATE_B3_BIT-EXACT_+_SE...	FAIL
V_04	BENCHMARK_GATE_B4_ENCODE_LATENCY	PASS
V_05	BENCHMARK_GATE_B5_DECODE_LATENCY	PASS
V_06	RED-TEAM_ATTACK_1_BASELINE	PASS
V_07	RED-TEAM_ATTACK_2_ENTROPY	PASS
V_08	RED-TEAM_ATTACK_3_LOSSLESS_QUALI...	FAIL
V_09	RED-TEAM_ATTACK_4_CORPUS_ADEQUACY	INC
V_10	RED-TEAM_ATTACK_5_FALSE-POSITIVE...	PASS
V_11	RED-TEAM_ATTACK_6_PYTHON_3.12_PA...	PASS
V_12	RED-TEAM_ATTACK_7_EXTERNAL_REPRO...	INC

Proof Anchors

Path	State
proofs/ENGINEERING_BLOCKERS.md	VERIFIED
proofs/narrow_claim/NARROW_CLAIM_GATE.json	VERIFIED
proofs/enterprise_benchmark/GATE_VERDICTS.json	VERIFIED
proofs/red_team/red_team_report.json	VERIFIED
proofs/release_candidate/clean_clone_result.json	VERIFIED
proofs/release_candidate/it04_parity_matrix_result.json	VERIFIED
proofs/imc_audit/imc_architecture_audit.json	VERIFIED

• Runtime and proof artifacts outrank prose.
• GO and NO-GO language is reserved for named gates only.
• Historical proof bundles remain lineage only. They do not override the March 21 blocker-state evidence or the later Phase 10 threshold evidence.
• No IMC runtime import is introduced by this repo.
• The current installable package artifact is zpe-robotics 0.1.0, but the repo's blocker-state docs remain the authority surface for engineering status.

Use these files together:

Need	Read first
Current blocker truth	proofs/ENGINEERING_BLOCKERS.md
Claim boundary	docs/CLAIM_BOUNDARY.md
Mechanics layer	docs/MECHANICS_LAYER.md
Benchmark verdicts	proofs/enterprise_benchmark/GATE_VERDICTS.json
Adversarial findings	proofs/red_team/red_team_report.json
Package/runtime boundary	proofs/runbooks/TECHNICAL_RELEASE_SURFACE.md
Release-candidate note	docs/RELEASE_CANDIDATE.md
Citation record	CITATION.cff
Docs registry	docs/DOC_REGISTRY.md
Historical lineage	proofs/README_LINEAGE_PATHS.md

Expanded LeRobot Benchmark Coverage

The expanded benchmark sweep is recorded in proofs/artifacts/lerobot_expanded_benchmarks/. The current qualified surface covers 3 real datasets across 3 materially distinct families, with compression ratios spanning 58.70x to 186.05x.

Dataset	Family	Compression ratio	Status
lerobot/columbia_cairlab_pusht_real	pusht	186.05x	qualified
lerobot/aloha_mobile_shrimp	aloha	61.27x	qualified
lerobot/umi_cup_in_the_wild	umi	58.70x	qualified
lerobot/pusht_image	pusht	n/a	skipped: insufficient joint dimension

Read proofs/artifacts/lerobot_expanded_benchmarks/aggregate_spread_summary.json for the rollup verdict and proofs/artifacts/lerobot_expanded_benchmarks/dataset_manifest.json for the full attempt ledger, including qualification misses that were preserved rather than averaged away.

Repo Shape

Field	Value
Proof Anchors	7
Modality Lanes	3
Authority Source	proofs/ENGINEERING_BLOCKERS.md

The modality-lane count reflects the three recorded parity lanes (arm64-qemu, macos, ubuntu-x86) in proofs/release_candidate/it04_parity_matrix_result.json.

Area	Purpose
src/zpe_robotics/	package implementation, CLI, packet handling, search, anomaly, and audit logic
tests/	release-surface, CLI, codec, and regression checks
scripts/	replay, benchmark, falsification, and clean-clone helpers
docs/	front-door, architecture, support, legal, and family-linkage docs
proofs/	blockers, benchmark artifacts, red-team outputs, release runbooks, and historical bundles
.github/workflows/	CI, clean-clone, parity, comparator, and publish workflows

Quick Start

Surface	Current truth
Repository	https://github.com/Zer0pa/ZPE-Robotics.git
Package / import / CLI	zpe-robotics / zpe_robotics / zpe-robotics
Acquisition surface	pip install zpe-robotics (available on PyPI)
License	LicenseRef-Zer0pa-SAL-7.0
Contact	[e-mail]
Release state	public repo and published package; engineering surface remains blocker-governed
Engineering	not complete
Current authority	proofs/ENGINEERING_BLOCKERS.md

Authority layer	File
governing blocker state	proofs/ENGINEERING_BLOCKERS.md
benchmark gate verdicts	proofs/enterprise_benchmark/GATE_VERDICTS.json
adversarial verdicts	proofs/red_team/red_team_report.json
package/runtime boundary	proofs/runbooks/TECHNICAL_RELEASE_SURFACE.md
IMC integration boundary	proofs/imc_audit/imc_architecture_audit.json

Install from PyPI:

pip install zpe-robotics
zpe-robotics --version

Or install from source (development):

pip install -e .
zpe-robotics --version

Repo-local engineering surface:

python -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip setuptools wheel
python -m pip install -e ".[dev,benchmark,telemetry,netnew]"
python -m pytest tests -q
python -m build

If you need the shortest honest verification route, use docs/AUDITOR_PLAYBOOK.md. If you need the release workflow boundary, use proofs/runbooks/TECHNICAL_RELEASE_SURFACE.md.

Need	Route
Security reporting	SECURITY.md
Claim boundary	docs/CLAIM_BOUNDARY.md
Support routing	docs/SUPPORT.md
Docs index	docs/README.md
Operator commands	docs/OPERATOR_RUNBOOK.md

Ecosystem

ZPE-Robotics follows the portfolio release-hygiene pattern used by ZPE-IMC, but it does not inherit IMC runtime claims, benchmark verdicts, or release readiness by association.

Need	Route
Robotics-to-IMC boundary	docs/family/ROBOTICS_RELEASE_LINKAGE.md
Frozen proof lineage note	proofs/README_LINEAGE_PATHS.md
Release-hygiene reference (descriptive only, not a runtime dependency)	ZPE-IMC

Observability: Comet dashboard (public)

Who This Is For

ZPE-Robotics is one of 17 independent encoding products in the Zer0pa portfolio — each domain-specific, each carrying its own proof surface. Robotics addresses motion telemetry; it does not inherit claims or release status from any other lane.

Ideal first buyer	Robotics infrastructure team or simulation/replay platform
Pain	Robot telemetry archives grow fast and can only be searched after full decompression — replay pipelines lack determinism guarantees
Deployment	Public Python package — pip install zpe-robotics
Family position	Motion-telemetry lane in the Zer0pa portfolio. Documentation/release-hygiene linkage to ZPE-IMC is descriptive only, not a runtime dependency; see docs/family/ROBOTICS_RELEASE_LINKAGE.md.