Cassie Model
Repo containing the urdf and frost based models of Cassie
This repository can be used to generate a model of the Cassie-series robot using [FROST](http://ayonga.github.io/frost-dev/). The project is written primarily in MATLAB, distributed under the BSD 3-Clause "New" or "Revised" License license, first published in 2017. Key topics include: cassie, frost, urdf-models.
Cassie Model
This repository can be used to generate a model of the Cassie-series robot using FROST.
Requirements
- MATLAB 2016b or later
- MATLAB supported compiler for generating MEX files
- Mathematica 11
- FROST
Generating Cassie's kinematics and dynamics
- Open the
model_gen.mscript in MATLAB. - Modify
FROST_PATHto reflect your FROST installation directory. - Generate the Cassie FROST model by runnning the first block of code, ending at line
cassie = Cassie('urdf/cassie.urdf'). - Run the rest of the script to generate and export various kinematic and dynamics functions.
Cassie coordinate system
Base DOFs
q(1)- xq(2)- yq(3)- zq(4)- yawq(5)- pitchq(6)- roll
Left side joints
q(7)- hip abductionq(8)- hip rotationq(9)- hip flexionq(10)- knee jointq(11)- knee spring (constrained to 0)q(12)- ankle joint (constrained to q12 = deg2rad(13) - q10)q(13)- toe joint
Right side joints
q(14)- hip abductionq(15)- hip rotationq(16)- hip flexionq(17)- knee jointq(18)- knee spring (constrained to 0)q(19)- ankle joint (constrained to q19 = deg2rad(13) - q17)q(20)- toe joint
Contributors
Showing top 2 contributors by commit count.
Related Repositories
rohanpsingh/LearningHumanoidWalking
Training a humanoid robot for locomotion using Reinforcement Learning
UMich-BipedLab/extrinsic_lidar_camera_calibration
This is a package for extrinsic calibration between a 3D LiDAR and a camera, described in paper: Improvements to Target-Based 3D LiDAR to Camera Calibration. This package is used for Cassie Blue's 3D LiDAR semantic mapping and automation.
UMich-BipedLab/LiDARTag
This is a package for LiDARTag, described in paper: LiDARTag: A Real-Time Fiducial Tag System for Point Clouds
UMich-BipedLab/CLF_reactive_planning_system
This package provides a CLF-based reactive planning system, described in paper: Efficient Anytime CLF Reactive Planning System for a Bipedal Robot on Undulating Terrain. The reactive planning system consists of a 5-Hz planning thread to guide a robot to a distant goal and a 300-Hz Control-Lyapunov-Function-based (CLF-based) reactive thread to cope with robot deviations. The planning system allowed Cassie Blue to autonomously traverse sinusoidally varying terrain. More experiments are still being conducted and this repo and the paper will be updated accordingly.