Page 102 - Read Online

Page 292
A1
A1
A1
A1
A1
ANYmal
Solo8
Unitree
Unitree
Unitree
Unitree
Unitree
Pybullt
Pybullet
Pybullet
RaiSim
RaiSim
RaiSim
RaiSim
Isaac,
Position,
Veloc-
Smooth-
End-
Orientation,
Ori-
Consump-
End-Effector,
Ve-
Tra-
Action
Linear and Angular Velocity, and
and Hip
Foot
Reward.
and
Velocities.
Height,
Motion,
Slip.
Velocities,
Motion,
Angular
Pose
Lateral Movement,
Collision,
Base
Impact,
Goal.
Foot
Energy
Hips
and
Base
and
and
Joint
and
Speed,
Imitation,
Tracking,
and
and
Joint
and
Velocities
Ground
Position,
State,
Motion
Gap,
Torques
Acceleration.
Knee
Orientation,
Quaternion
Gaits,
States
Joint
entation,
Velocity
Effector
jectory
Work,
Joints.
locity.
State,
Rate,
ness,
tion,
Joint
Joint
Joint
Joint
Base
Base
ity.
Ve-
End-
Joint
Joint
Joint
Joint
Joint
Deviation.
and
Torques.
Deviation,
State
Positions [43]
Positions.
Positions.
Positions.
Positions.
Positions.
Desired
Desired
Effector
Desired
Desired
Desired
Desired
locities
locity
Ex-
Ac-
Ve-
Orientation, Foot Contact, Previ-
Po-
Po-
An-
(3-step),
Velocities,
Velocities,
Velocities,
Command,
and
Height
Foot
and
and
Joint
history),
Trajectory
Command
Roll)
Joint
and
(Yall
Action
Gravity,
Velocities,
Actions
and
(3-step),
(2-dim),
and
and
and
history).
Velocity,
Poses.
Velocities
(with
(Yall
Previous
Corrected
Vector.
Proprioception,
Positions
Velocities,
Positions
Positions
Vector.
(3-step),
Target
Data
Action.
and
Orientation
Orientation
Orientation
Action.
(with
and
Contact
Angular
locities,
Sensor
Future
sitions
trinsic
Maps.
Angle Base Position, Goal BaseTargetPositions(withPrevi- and Orientation, Error, Yobogo Webots [106] and Pitch, Velocity, Survival. and Orientation, ous), Command. Height. Max. Lift Leg Direction, Goal Map, Elevation and State Joint Velocities, Base Velocity, Angular and Linear Frequen- Phases, FTG Velocity, Foot Collision, and Motion Base Joint Desired A1 Unitree Pybullet His- Joint FTG), and (Base cies Smoothness, Target Clearance, Positions. Con- and Targets Foot tory, Map. Traversability and Torque, Env. States, Contact Forces, tact Param. Velocities, and Positio
sition
Poses
Roll).
Joint
Joint
Joint
Base
ities,
Last
gles
ous
REDQ [29]
SAC SAC PPO PPO PPO PPO PPO PPO PPO
quadruped lo- learning terrains con- DRL-based stability. noisy get to to order in tracking propri- and of tasks locomo- to related is real-time for in problems control motion opera- human control torque high- via torques robust create us- training demon- fine- for locomotion
well-performing for system real-world in pre-training. terrain-aware troller integrating a RAN to guar- action the DRL using trajectory quadrupedal a vision Incorporating navigation in robots. Energy constraints leading to the natural of choice the and speed. desired algorithm adaptation robots. quadrupedal allowing quadrupedal predicting joint to approach policiesdeployableonrealrobots additional optimised single system RL robot real-world
A robot comotion without A antee policy A reference generate system. oception legged emergence tion, the RMA online quadruped A system tion. A framework frequency RL. RL A without a ing stration. A tuning policies.
2021 2021 2021 2021 2021 2021 2022 2022 2022 2022
IROS IROS ICRA CVPR workshop CoRL RSS RSS ArXiv ArXiv ICRA
for on for Tough Adaptation using in Gaits for of DRL [108] for Learn- Policies
Framework Based Locomotion Risk-Assessment- DRL (RAN) in Locomotion Locomotion Propriorception and Robots [11] Legged Consumption of Emergence Adaptation Control using Control Locomotion [44] Model-freeRLforRobustLocomotion usingDemonstrationsfromTrajectory on Keep Locomotion
Hierarchical Terrain-Aware Network-Aided Quadrupedal Trajectory Quadrupedal Vision of navigation Energy the to Robots [107] Motor Rapid Robots [60] Motion Robots Quadrupedal Torque Quadrupedal Optimization [109] that Robots Fine-Tuning World [30] Real
A Quadruped RL [105] Terrain [51] Real-Time for DRL [53] Coupling for Minimising Leads Legged RMA: Legged Human Learning Legged ing: the in