SkateNav: Learning Energy-⁠Efficient Navigation
through Adaptive Skating on Passive-⁠Wheel Quadrupedal Robots

Abstract

Passive-wheel quadrupedal robots offer energy-efficient locomotion through leg-driven propulsion and momentum-driven gliding. However, navigation requires coordinating steering with intermittent propulsion and adapting push–glide timing to task demands. Velocity-tracking objectives may suppress the speed variations inherent in skating, discouraging efficient gliding. To address these challenges, we propose SkateNav, a unified skating-navigation framework based on reinforcement learning. A time-budgeted goal-reaching objective allows the policy to jointly learn steering, propulsion, and their temporal allocation. An adaptive skating cycle structures gliding, pushing, and recovery, while frequency modulation at every control step adjusts phase durations to navigation demands. Simulation experiments demonstrate higher average success rates and lower cost of transport than the evaluated planner–controller baselines, with ablations confirming the contributions of cycle structure and frequency adaptation. Real-world experiments demonstrate zero-shot transfer from simulation to a passive-wheel Mini Cheetah across cluttered indoor spaces, slopes, and paved outdoor surfaces.

1. 🎯
   Problem formulation We formulate passive-wheel local navigation as a time-budgeted goal-reaching problem in which steering, propulsion, and push–glide timing are learned jointly.
2. 🌀
   Adaptive skating cycle We introduce an adaptive skating cycle representation that preserves leg coordination while exposing phase progression as a navigation-conditioned policy action.
3. 📊
   Evaluation We demonstrate higher average success rates and lower cost of transport than the evaluated planner–controller baselines, conduct component ablations, and validate real-world navigation across indoor and outdoor environments.

Experimental Evaluation

Simulation · five conditions · 1000 navigation tasks per condition

Navigation performance in simulation environments

Best Second-best results are highlighted within each comparison group, with Ours included in both. For w/o Adaptive Freq., we report results at the fixed frequencies yielding the highest SR: 2.0 Hz (Flat/BARN-easy), 1.8 Hz (BARN-hard/Uphill), and 1.2 Hz (Downhill). For planners, we select the best cost setting (aggressive, balanced, or safe) by average SR: balanced for DWA; aggressive for MPPI and CEM.

Ours achieves the highest Success Rate (SR) and the lowest Cost of Transport (CoT) and Inference Time (IT), outperforming the planner–VT baselines. For Flat, all methods achieve high SR, whereas performance differences become more pronounced as navigation demands increase.

Adaptive skating frequency

As the waypoint distance increases under the same time budget, the policy raises the skating frequency, producing more frequent Push phases. This tendency is strongest on Uphill, where repeated foot-ground contact provides traction for traversal, whereas Downhill favors lower frequencies to exploit momentum gained during descent. Furthermore, the skating frequency increases as obstacle clearance decreases in the BARN settings, enabling rapid steering corrections.

Temporal allocation of pushing and gliding

VT allocates more of each cycle to Glide–Push and less to Glide–Recovery than Ours, possibly to track the commanded velocity. In contrast, Ours travels comparable distances during G-P and G-R while consuming less energy in G-R than in G-P. This temporal redistribution toward momentum-driven motion supports the consistently lower CoT of Ours compared with the planner–VT baselines.

Skating cycle component ablation

The skating frequency required for favorable performance varies with navigation demand. Uphill requires sufficiently high frequencies for reliable navigation, whereas higher frequencies can raise CoT on Flat and BARN-hard. Thus, a single fixed frequency may not achieve both high SR and low CoT across varying navigation demands.

Without the skating cycle, the goal-reaching objective alone may struggle to discover an energy-efficient skating pattern. w/o Skating Cycle yields lower SR and substantially higher CoT across all environments, because the learned motion involves sustained wheel–ground contact accompanied by persistent slip.

BibTeX

Venue will be filled in on publication

@inproceedings{skatenav,
  title     = {{SkateNav}: Learning Energy-Efficient Navigation through
               Adaptive Skating on Passive-Wheel Quadrupedal Robots},
  author    = {Jeong, Jeil and Kim, Taeyun and Kim, Sangjin and Choi, Jinyoung and Yoon, Sung-Eui},
  booktitle = {TBD},
  year      = {TBD}
}