PyRobot: An Open-source Robotics Framework for Research and Benchmarking

By Adithyavairavan Murali, Tao Chen, Kalyan Vasudev Alwala, Dhiraj Gandhi, Lerrel Pinto, Saurabh Gupta, Abhinav Kumar Gupta. 2019

118 citations recorded by the source corpus.

This paper introduces PyRobot, an open-source robotics framework for research and benchmarking. PyRobot is a light-weight, high-level interface on top of ROS that provides a consistent set of hardware independent mid-level APIs to control different robots. PyRobot abstracts away details about low-level controllers and inter-process communication, and allows non-robotics researchers (ML, CV researchers) to focus on building high-level AI applications. PyRobot aims to provide a research ecosystem with convenient access to robotics datasets, algorithm implementations and models that can be used to quickly create a state-of-the-art baseline. We believe PyRobot, when paired up with low-cost robot platforms such as LoCoBot, will reduce the entry barrier into robotics, and democratize robotics. PyRobot is open-source, and can be accessed via this https URL.

Subfields: Control and Systems Engineering, Mechanical Engineering, Reinforcement Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

In Mechanical Engineering

In Reinforcement Learning

Citations in the corpus

Experience Grounds Language (2020, background): “, 2020) and the cost and availability of commodity hardware (Fitzgerald, 2013; Campeau-Lecours et al., 2019; Murali et al., 2019).”
AI2-THOR: An Interactive 3D Environment for Visual AI (2017, background): “ManipulaTHOR [5] StretchRE1 [14] LoCoBot [24] Abstract Drone [42] Figure 4: The current agents available in AI2-THOR include the ManipulaTHOR and StretchRE1 agents, which support arm manipulation, and the LoCoBot, Abstract, and Drone agents, which support navigation and abstracted interaction.”
Habitat: A Platform for Embodied AI Research (2019, methodology): “To model this, we introduce a noise model acquired by benchmarking a real robotic platform [22].”
SAPIEN: A SimulAted Part-Based Interactive ENvironment (2020, background): “2) The environment should incorporate the simulation of real robots, being able to reproduce the behaviors of real robotics manipulators, sensors and controllers [34].”
Differentiable SLAM-net: Learning Particle SLAM for Visual Navigation (2021, methodology): “Unless stated otherwise, we use the Habitat Challenge 2020 configuration [32]: Gaussian noise for RGB images; the Redwood noise model for depth images [13]; the Locobot motion model for actuation noise [48].”
Learning to Explore using Active Neural SLAM (2020, methodology): “We use the pyrobot API (Murali et al., 2019) along with ROS (Quigley et al., 2009) to implement the control commands and get sensor readings.”
SoundSpaces: Audio-Visual Navigation in 3D Environments (2019, background): “The API offers fast (over 10K fps) rendering and support for multiple datasets [83,104,12,66,21].”
Efficient Bimanual Manipulation Using Learned Task Schemas (2019, unknown): “We control the Sawyers using PyRobot [35].”
Audio-Visual Embodied Navigation (2019, background): “The API offers fast (over 10K fps) rendering and support for multiple datasets [85, 108, 14, 68, 23].”
Object Goal Navigation using Goal-Oriented Semantic Exploration (2020, methodology): “We used the Locobot hardware platform and PyRobot API [32] to deploy the trained policy in the real world.”
Cross-View Semantic Segmentation for Sensing Surroundings (2019, methodology): “To verify the performance of our model in the real-world robotic environment, we conduct a semantic navigation experiment by using a LoCoBot mobile robot [20] (cf.”
TriFinger: An Open-Source Robot for Learning Dexterity (2020, background): “More recently Murali et al. (2019b) proposed PyRobot, a python interface for motion generation and learning.”
RoboTHOR: An Open Simulation-to-Real Embodied AI Platform (2020, methodology): “More recent efforts have led to low cost robot solutions [58, 3, 43] opening up the space to more research entities.”
Sim2Real Predictivity: Does Evaluation in Simulation Predict Real-World Performance? (2019, methodology): “To model the actions on LoCoBot, we leverage an actuation noise model derived from mocap-based benchmarking by the PyRobot authors [18].”
Occupancy Anticipation for Efficient Exploration and Navigation (2020, methodology): “It is trained to minimize the pixelwise (cid:96) 1 loss between the prediction and the ground-truth. in the robot motions is simulated by using a noise model obtained from the LoCoBot [43].”
Are We Making Real Progress in Simulated Environments? Measuring the Sim2Real Gap in Embodied Visual Navigation (2019, methodology): “For modeling the actuation noise in the simulator we used data from [18] and used a Gaussian distribution to fit a model on the data.”