Sim2Real Predictivity: Does Evaluation in Simulation Predict Real-World Performance?

TL;DR

This paper investigates how well simulation-based evaluations of robot navigation predict real-world performance, introducing tools and metrics to measure and improve this sim2real transferability.

Contribution

The authors develop the Habitat-PyRobot Bridge for seamless simulation-to-robot transfer and introduce the SRCC metric to quantify sim2real predictivity, demonstrating how tuning simulation parameters enhances transfer success.

Findings

Low SRCC (0.18) indicates poor predictivity in unadjusted simulation.

Agents exploit simulation imperfections, reducing real-world applicability.

Tuning simulation parameters significantly improves SRCC to 0.844.

Abstract

Does progress in simulation translate to progress on robots? If one method outperforms another in simulation, how likely is that trend to hold in reality on a robot? We examine this question for embodied PointGoal navigation, developing engineering tools and a research paradigm for evaluating a simulator by its sim2real predictivity. First, we develop Habitat-PyRobot Bridge (HaPy), a library for seamless execution of identical code on simulated agents and robots, transferring simulation-trained agents to a LoCoBot platform with a one-line code change. Second, we investigate the sim2real predictivity of Habitat-Sim for PointGoal navigation. We 3D-scan a physical lab space to create a virtualized replica, and run parallel tests of 9 different models in reality and simulation. We present a new metric called Sim-vs-Real Correlation Coefficient (SRCC) to quantify predictivity. We find that SRCC for Habitat as used for the CVPR19 challenge is low (0.18 for the success metric), suggesting that performance differences in this simulator-based challenge do not persist after physical deployment. This gap is largely due to AI agents learning to exploit simulator imperfections, abusing collision dynamics to 'slide' along walls, leading to shortcuts through otherwise non-navigable space. Naturally, such exploits do not work in the real world. Our experiments show that it is possible to tune simulation parameters to improve sim2real predictivity (e.g. improving $SRCC_{Succ}$ from 0.18 to 0.844), increasing confidence that in-simulation comparisons will translate to deployed systems in reality.