Organizations: Institute of Humanoid Robots, Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China, Hefei, Anhui 230026, China.
Although recent years have seen significant progress of humanoid robots in walking and running, the frequent foot strikes with ground during these locomotion gaits inevitably generate high instantaneous impact forces, which leads to exacerbated joint wear and poor energy utilization. Roller skating, as a sport with substantial biomechanical value, can achieve fast and continuous sliding through rational utilization of body inertia, featuring minimal kinetic energy loss. Therefore, this study proposes a novel humanoid robot with each foot equipped with a row of four passive wheels for roller skating. A deep reinforcement learning control framework is also developed for the swizzle gait with the reward function design based on the intrinsic characteristics of roller skating. The learned policy is first analyzed in simulation and then deployed on the physical robot to demonstrate the smoothness and efficiency of the swizzle gait over traditional bipedal walking gait in terms of Impact Intensity and Cost of Transport during locomotion. A reduction of 75.86% and 63.34% of these two metrics indicate roller skating as a superior locomotion mode for enhanced energy efficiency and joint longevity.
Figures & tables
Fig. 1: Overview of the SKATER platform and its learned swizzling gait.
Fig. 2: Hardware design of SKATER and its passive-wheel inline-skate foot module.
Fig. 3: Deep reinforcement learning control framework for SKATER. The policy network processes proprioceptive sensor data to generate joint-level commands, enabling adaptive inline skate swizzles.
Reward Term
Formula
Weight
Task Rewards
Alive
1alive
0.15
Linear Velocity Track
exp(−0.25∥vxycmd−vxyyaw∥2)
1.8
Angular Velocity Track
exp(−0.25(ωzcmd−ωzbase)2)
1.2
Joint Penalties
Arms Deviation
∑arms∣qi−qi0∣
−0.4
TABLE I: Reward Function Terms
Term
Value
Material Physics - Wheels
Static Friction Coefficient
U(0.1,0.8)
Dynamic Friction Coefficient
U(0.1,0.4)
Base Mass Offset
U(−1.0,3.0) kg added to pelvis
CoM Offset - Pelvis and Torso
x/y/z axes
U(−0.01,0.01) m
TABLE II: Domain Randomization Parameters
Fig. 4: Comparison of vertical GRF profiles.
Fig. 5: Velocity tracking performance of SKATER following step commands.
Fig. 6: Temporal analysis of the learned swizzling gait: ankle joint angle profiles and inter-foot distance variation with the reward bounds indicated.
(a) Main performance metrics
Method
Task Survival (%) ↑
Worst-cmd. Survival (%) ↑
Fall Rate (%) ↓
Forward Vel. RMSE (m/s) ↓
Yaw-rate RMSE (rad/s) ↓
Energy per Meter ↓
Full reward
94.62 ± 0.75
64.80 ± 9.09
5.38 ± 0.75
0.236 ± 0.001
0.194 ± 0.001
113.95 ± 0.27
w/o symmetry reward
99.17 ± 0.41
93.00 ± 3.08
0.83 ± 0.41
0.257 ± 0.000
0.329 ± 0.000
140.13 ± 0.11
w/o ZMP reward
87.22 ± 1.13
26.40 ± 2.70
12.78 ± 1.13
0.196 ± 0.001
0.255 ± 0.002
81.91 ± 0.61
w/o ZMP margin reward
90.00 ± 0.93
0.00 ± 0.00
10.00 ± 0.93
0.241 ± 0.003
0.251 ± 0.000
90.92 ± 0.84
w/o foot-distance reward
86.19 ± 0.93
0.00 ± 0.00
13.81 ± 0.93
0.273 ± 0.001
0.326 ± 0.004
154.34 ± 1.15
TABLE III: Reward term ablations. Values are reported as mean ± standard deviation over five random seeds.
Fig. 7: Horizontal foot trajectories under representative reward term ablations and their motion-level failure modes.
As humanoid robots become increasingly dynamic, coupling them with reinforcement learning offers a promising approach to solving the complex, underactuated mechanics of passive inline skating. Equipping a humanoid robot with passive inline skating wheels presents an opportunity to combine the versatile agility of humanoids with the high-speed, energy-efficient locomotion strategies utilized by human skaters. In this paper, we train and deploy a reinforcement learning control policy that enables novel locomotion strategies for a humanoid robot modified to equip consumer inline skates instead of conventional feet. Unlike previous work limited to quadrupedal robots or actively driven wheels, our system allows for precise 6-DoF control of the skates to execute dynamic, edge-driven propulsion strategies. Our skating strategies emerge entirely from our reward structure, without reliance on human motion data, imitation learning, or kinematic priors. We overcome the inherent instability of passive wheels and simulation contact artifacts by utilizing different geometric wheel models (spherical and ellipsoidal) during training and validation, along with a custom success-based command curriculum and a specialized rolling reward. Consequently, our policy demonstrates up to a 50% reduction in Cost of Transport (CoT) compared to standard walking gaits. The resulting policy successfully transfers zero-shot to the physical Booster T1 hardware. Real-world deployments demonstrate dynamic balance, the ability to reject active physical perturbations, and agile locomotion strategies capable of turning at speed. A video of our results can be found at https://www.youtube.com/watch?v=-_APcOS7uFo.
Ethan Marot, Thomas Bi, Clemens Schwarke +3
Institute for Dynamic Systems and Control, ETH Zurich, Switzerland. · Robotic Systems Lab, ETH Zurich, Switzerland
In this paper, we present a hardware-control co-design approach that enables efficient and versatile roller skating on quadrupedal robots equipped with passive wheels. Passive-wheel skating reduces leg inertia and improves energy efficiency, particularly at high speeds. However, the absence of direct wheel actuation tightly couples mechanical design and control. To unlock the full potential of this modality, we formulate a bilevel optimization framework: an upper-level Bayesian Optimization searches the mechanical design space, while a lower-level Reinforcement Learning trains a motor control policy for each candidate design. The resulting design-policy pairs not only outperform human-engineered baselines, but also exhibit versatile behaviors such as hockey stop (rapid braking by turning sideways to maximize friction) and self-aligning motion (automatic reorientation to improve energy efficiency in the direction of travel), offering the first system-level study of dynamic skating motion on quadrupedal robots.
Hanwen Wang, Zhenlong Fang, Josiah Hanna +1
Hanwen Wang, Zhenlong Fang, and Josiah Hanna are with the University of Wisconsin - Madison. · Xiaobin Xiong was with the University of Wisconsin - Madison, and now with the Shanghai Innovation Institute (SII).
Humanoid roller-skating is difficult because the robot must coordinate whole-body balance, rolling contacts, and velocity-dependent posture regulation. This paper presents an adversarial motion prior based reinforcement learning framework for two humanoid roller-skating gaits: Pump Glide skating and Push Glide skating. The two gait datasets are collected independently through motion capture and retargeted to the humanoid robot separately. The retargeted data are then smoothed and resampled into reference motion states for AMP training. The two gaits are learned by independent AMP training pipelines with separate reference datasets, separate policies, and independent reward architectures. Simulation experiments are designed to evaluate gait quality, velocity tracking, turning, and gait-specific reward ablations.
Yunkang Cheng, Yutong Wu, Menghan Li +2
Department of Automation, Tsinghua University, Beijing, China · Beijing Key Laboratory of Embodied Intelligence Systems, Beijing, China · Institute for Embodied Intelligence and Robotics, Tsinghua University, Beijing, China