Skip to content
Log in

Similar papers

Review Aug 2026

Evolution of humanoid locomotion control.

Humanoid robots stand at the forefront of robotics, aiming to capture the agility, robustness, and expressivity of human movement in an anthropomorphic form. The locomotion control of humanoids has evolved from classical model-based methods to reinforcement learning powered by large-scale simulation and now to generative models that produce adaptive, whole-body behaviors, propelling humanoids toward operation in real-world environments. This survey positions humanoid control at a turning point, converging toward a unified paradigm of physics-guided generative intelligence that integrates optimization, learning, and predictive reasoning. We identify three core principles linking these paradigms: physics-based modeling, constrained decision-making, and adaptation to uncertainty. Building on these connections, we provide recommendations for researchers and outline open challenges in safety, accessibility, and human-level capability. These directions represent a transformation from engineered stability to intelligent autonomy, laying the groundwork for humanoid generalists capable of operating safely, collaborating naturally, and extending human capability in the open world.

Yan Gu, Guanya Shi, Fan Shi et al. · 7 citations
Conference Jul 2026

Human-Inspired Redundancy Resolution for Upper-Limb Robotics Using Phase-Dominant Performance Criteria

The aim of this study is to develop improved motion planning algorithms for humanoid robots and to enhance prosthetic control and rehabilitation training through predictive interfaces grounded in natural human movement. To address this aim, we introduce a synthesized biomechanically plausible human-inspired inverse-kinematics framework for a 10-DOF Robotic Human Upper-Body Model (RHUBM) that resolves redundancy via phase-dominant performance criteria. We hypothesize that humans prioritize lighter joint movements over heavier ones and integrate various performance criteria such as manipulability, velocity ratio, and mechanical advantage while avoiding joint limits during different Activities of Daily Living (ADLs) tasks and propose a phase-dependent weighting strategy derived from statistical analysis of human motion. To validate our hypothesis, a subject-specific motion-capture dataset of Range-of-Motion and ADL tasks was created, phase-dominant performance criteria were identified by segmenting each task trajectory into major phases, and a Weighted Least-Norm (WLN) inverse kinematics controller was implemented whose joint-space weight matrix combines link weighting, joint-limit avoidance weights, and gradient-based weights derived from the phase-dominant performance criteria. The WLN algorithm outcomes were compared against motion capture (MoCap) data and the Least Norm (LN) solution. Results on a representative ADL (Drinking) demonstrate phase-dependent criterion switching, consistent with our hypothesis: Linear manipulability dominates during grasp/return phases, while angular velocity ratio dominates near the mouth. Across joints, WLN produces MoCap-consistent trajectories with consistently lower RMS errors than the LN solution. The framework formalizes human-inspired redundancy resolution and supports principled motion synthesis for humanoids, prosthetics, and rehabilitation interfaces.

Urvish Trivedi, Dimitrios Menychtas, Redwan Alqasemi et al. · 0 citations
Conference Jul 2026

Reinforcement Learning-based Impact Forces-Optimized Gait Control for Humanoid Locomotion: Toward Smooth and Quiet Walking

This paper proposes a reinforcement learning-based Impact Forces-Optimized Gait Control framework for achieving smooth and quiet humanoid locomotion. The proposed method integrates a quintic polynomial swing-foot trajectory designed to ensure zero vertical velocity and acceleration at foot-ground contact with a joint-jerk minimization reward. Impact forces are provided as privileged observations during training to improve robustness under an asymmetric information architecture. The effectiveness of the proposed method was verified through evaluations conducted in Isaac Lab and MuJoCo (Multi-Joint dynamics with Contact), demonstrating reduced impact forces across various locomotion modes. Additionally, velocity tracking evaluations confirm that the proposed method does not significantly degrade locomotion performance. Furthermore, its performance was validated on the G1 humanoid robot, where real-world walking tests confirmed a reduction in walking-induced acoustic noise.

Hyeonseok Jeong, Yunsoo Kim, Changeui Shin et al. · 0 citations
Conference Jul 2026

Underactuated Virtual Gravity Control: Harnessing Passive Dynamics for Optimally Efficient Bipedal Locomotion

This paper explores the Underactuated Virtual Gravity (UVG) controller, a model-based control approach designed to achieve energy-efficient bipedal walking. The UVG controller minimizes actuator effort during level-ground walking by capitalizing on the inherent dynamics that facilitate stable passive gaits on downward slopes. By effectively leveraging torso dynamics to support the application of the UVG, the method turns the innate underactuation of bipedal systems into an advantage. Extensive benchmarking against state-of-the-art methods such as the Trajectory Optimization for trajectory planning combined with Non-linear Model Predictive Control for tracking shows that the UVG achieves superior energy efficiency within its effective range while constituting a closed-form controller that requires fewer computational resources than numerical methods. The results highlight the UVG and related dynamics-based approaches as a compelling option for energy-conscious, task-focused robot designs.

Aikaterini Smyrli, E. Papadopoulos · 0 citations
Preprint Aug 2026

Closing the Loop on the Poppy Humanoid: Bipedal Locomotion with Linear-Quadratic Control and Learned Cost Functions

The Poppy Humanoid is an open-source, low-cost robot suitable for research and education in artificial intelligence. However, we are unaware of any published methodology that achieves reliable, unassisted bipedal locomotion on the standard Poppy hardware. This paper contributes a functional closed-loop walking controller for Poppy, based on the linear-quadratic regulator (LQR) framework for trajectory tracking. Starting with data collected from open-loop playback of a nominal walking trajectory, our proposed method learns a quadratic cost function for an LQR controller that substantially improves the reliability of the motion. The closed-loop controller is validated empirically, demonstrating statistically significant improvements in walking performance compared to open-loop trajectory playback.

Xulin Chen, Borui He, Rui-Peng Liu et al. · 0 citations
Open access Aug 2026

Locomotion Control Strategy Design and Simulation of Parallel-Legged Insect-Scale Micro Crawling Robot

High nonlinearity and limited computational resources create persistent challenges for achieving autonomous, stable, and accurate locomotion in insect-scale crawling robots, hindering their practical deployment. In this study, we build the mathematical models of the insect-scale micro crawling robot named PLioBot and propose a locomotion control strategy that eliminates the need for gait transitions. The locomotion transformation of the PLioBot prototype, from driving inputs to mechanical motion outputs, is decomposed into multiple motion processes. This study analyzes the theoretical models underlying these motion processes and develops the locomotion simulation model for the PLioBot based on the mathematical models and the multibody dynamics simulation tool. The locomotion control strategy with low computational requirements is designed based on a closed-loop PID controller, which regulates the robot’s locomotion by independently adjusting the step lengths of its left and right legs. The locomotion co-simulation system is established to validate the control strategy. The simulation results confirm that this control strategy enables the PLioBot to perform straight-line locomotion and turning without requiring any gait transition.

Qunwei Zhu, Tao Jiang, Zirong Luo et al. · 0 citations