ICRA Experiment Design & Audit
Audits and optimizes robotics experimental setups, baseline fairness, sim-to-real claims, and statistical reporting for ICRA paper submissions.
Designed for robotics researchers submitting to ICRA and premier robotics venues, this skill provides end-to-end guidance and auditing for paper experimental sections. It aligns research claims with empirical evidence across a 5-rung ladder (from simulation to field deployments), enforces statistical rigor tailored for small-sample robotics trials (n=10-25), ensures fair hardware baseline benchmarking, structures transparent sim-to-real transfer disclosures, and builds comprehensive failure-mode taxonomies that withstand reviewer scrutiny.
主要功能
01Trial count, reset protocol, and exact binomial statistical confidence auditing
02First-class failure taxonomy structuring and hardware ablation allocation budgeting
03Evidence ladder verification matching claim altitude to empirical reality (simulation vs. hardware)
04Sim-to-real transfer quantification and zero-shot vs fine-tuning delta tracking
05Hardware baseline benchmarking protocols ensuring fair tuning and classical comparisons
061,079 GitHub stars
使用场景
01Designing physical robot and simulation experiment protocols with appropriate sample sizes and abort rules
02Constructing structured failure analysis tables and hardware baseline comparison sections
03Auditing an ICRA paper draft before submission to eliminate claim-evidence mismatches