Load-Aware Locomotion Control for Humanoid Robots in Industrial Transportation Tasks

Lequn Fu; Yijun Zhong; Xiao Li; Yibin Liu; Zhiyuan Xu; Jian Tang; Shiqi Li

cs.ROMar 15, 2026

Load-Aware Locomotion Control for Humanoid Robots in Industrial Transportation Tasks

Authors: Lequn Fu, Yijun Zhong, Xiao Li, Yibin Liu, Zhiyuan Xu, Jian Tang, Shiqi Li

Organizations: Huazhong University of Science and Technology, Wuhan 430074, China · Beijing Humanoid Robot Innovation Center, Beijing 100086, China

Abstract

Humanoid robots deployed in industrial environments are required to perform load-carrying transportation tasks that tightly couple locomotion and manipulation. However, achieving stable and robust locomotion under varying payloads and upper-body motions is challenging due to dynamic coupling and partial observability. This paper presents a load-aware locomotion framework for industrial humanoids based on a decoupled yet coordinated loco-manipulation architecture. Lower-body locomotion is controlled via a reinforcement learning policy producing residual joint actions on kinematically derived nominal configurations. A kinematics-based locomotion reference with a height-conditioned joint-space offset guides learning, while a history-based state estimator infers base linear velocity and height and encodes residual load- and manipulation-induced disturbances in a compact latent representation. The framework is trained entirely in simulation and deployed on a full-size humanoid robot without fine-tuning. Simulation and real-world experiments demonstrate faster training, accurate height tracking, and stable loco-manipulation. Project page: https://lequn-f.github.io/LALO/

Figures & tables

Category	Dimension	Noise Scale
Phase Input	2	0
Projected gravity	3	$\pm 0.05$
Angular velocity	3	$\pm 0.1$
Joint positions	27	$\pm 0.02$
Joint velocities	27	$\pm 0.5$
Command	4	0

TABLE I: Observation space of the locomotion policy.

Fig. 1: Overview of the proposed load-aware humanoid loco-manipulation framework. Upper-body manipulation is generated by perception-driven kinematic control using 6D object pose estimation, while lower-body locomotion is governed by a residual reinforcement learning policy with height-conditioned offsets and history-based state estimation. Historical proprioceptive observations are encoded by a history-based state estimator to infer base velocity, base height, and a compact latent state capturing load- and manipulation-induced disturbances.

Reward term	Formulation	Weight $(\mu_{i})$
Linear velocity tracking	$\exp\!\left(-4\\|[v_{x},v_{y}]-[v_{x}^{\ast},v_{y}^{\ast}]\\|^{2}\right)$	$1.2$
Angular velocity tracking	$\exp\!\left(-4(\omega_{z}-\omega_{z}^{\ast})^{2}\right)$	$1.1$
Base height tracking	$\exp(-50\,\|h_{b,rel}-h_{b,rel}^{\ast}\|)$	$2.0$
Low-speed consistency	$r_{\mathrm{ls}}(v_{x},v_{x}^{\ast})\cdot\mathbb{I}(\|v_{x}^{\ast}\|>0.1)$	$0.2$
Orientation stabilization	$\exp\!\left(-10\,\\|g_{proj,xy}\\|\right)$	$1.0$
Velocity mismatch	$0.5(\exp(-10v_{z}^{2})+\exp(-5\\|\omega_{x,y}\\|))$	$0.5$

TABLE II: Reward terms and weights used for training.

Parameter	Unit	Range	Operator
Ground Friction	–	[0.1, 3.0]	–
Restitution Coefficient	–	[0.0, 1]	–
External Push Velocity (XY)	m/s	[-0.5, 0.5]	additive
External Push Angular Velocity	rad/s	[0, 0.6]	additive
Base Mass Offset	kg	[-5.0, 10.0]	additive
Hand Mass Offset	kg	[-0.1, 0.3]	additive

TABLE III: Overview of Domain Randomization

Fig. 2: Multi-box detection and target 6D pose estimation. Blue boxes denote instances detected by YOLO-seg, the red box indicates the selected target, and the green wireframe shows the 6D pose estimated by FoundationPose.

Parameter Name	Value
Value loss coefficient	1.0
Number of Epochs	5
Clipping	0.2
Entropy Coefficient	0.003
Discount Factor	0.99
GAE Discount Factor	0.95

TABLE IV: PPO Training Hyperparameters

Fig. 3: Training convergence comparison of different offset and reference formulations. (a) Episode reward for tracking base height versus environment steps. (b) Mean episode length versus environment steps.

Fig. 4: Distribution of joint residual actions during steady-state walking.

Fig. 5: Prediction performance of the history-based state estimator. Comparison between commanded, ground-truth, and estimated states for (a) base velocity

v_{x}

, (b) base velocity

v_{y}

, and (c) base height. The estimator achieves low mean squared error under dynamic motion conditions.

Fig. 6: Real-world loco-manipulation experiments in an industrial box transportation scenario. (a–c) Box transportation without depalletizing: the robot transitions from unloaded walking to load-carrying locomotion, squats to a target height, places the box onto a rack while maintaining a crouched posture, and then stands up. (d–f) Transportation with depalletizing: the robot performs box depalletizing, walks while carrying a $5$ kg payload, places the box at the designated location, and continues unloaded walking for the next task. The environment is a one-to-one reconstruction of an industrial factory setting and corresponds to the task scenario of the World Humanoid Robot Games.

Humanoid Loco-Manipulation

Humanoid loco-manipulation studies how robots coordinate walking, balance, and whole-body interaction to move and handle objects.

See all 86 papers

HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids
Oct 6, 2026An Dang, Arturo Flores Alvarez, Yu-Ming Chen +5Humanoid Loco-Manipulation Whole-Body Control
Amazon · University of Michigan · University of California, Los Angeles +1
Humanoid Rickshaw Pulling: Whole-Body Locomotion under Coupled Wheeled Loads
Oct 3, 2026Yangzhi Yang, Xiansheng Lin, Zhaoming Xie +1Humanoid Robot Control Humanoid Robot Locomotion
Legged AI Lab, Shanghai Innovation Institute
Dataset-Free Compliant Humanoid Loco-Manipulation with Dynamic Online Posture
Oct 5, 2026Seungho Yeom, Zhenyu Wu, Jaeyoung Huh +4Robotic Control Compliant Robot Control
University of California San Diego, La Jolla, CA 92093, USA. · Yonsei University, Seoul 03722, Republic of Korea.

Oct 6, 2026J/K move · Enter open · S save

HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids

An Dang, Arturo Flores Alvarez, Yu-Ming Chen, Conor Mc Gartoll+4

Amazon · University of Michigan · University of California, Los Angeles+1

Humanoid Loco-Manipulation Whole-Body Control Force-Controlled Robotic Manipulation RL Control

Open paper AskPDF arXiv

Humanoid loco-manipulation of large, heavy objects demands forceful interaction across the entire body. However, such payloads shift a humanoid's center of mass and impose sustained loads across the upper body, challenging balance and command tracking. We present HULK, a whole-body control framework for forceful loco-manipulation. Using model predictive control (MPC) to guide reinforcement learning with predictions of the loaded dynamics, we train two teachers: one tracks arm motions under wrist forces, and the other locomotes while holding large objects against the body. A capture-point control barrier function augments the wrist-force teacher during training to improve balance under load. We distill both teachers into a single policy. Evaluation spans simulation and the Unitree G1. In simulation, the teacher with the barrier function achieves the lowest forward and lateral velocity tracking errors at 10 kg per arm among evaluated controllers and reduces aggregate divergent component of motion (DCM) excursion magnitude by 35.7% relative to MPC-guided reinforcement learning alone. Our wrist-force teacher withstands torso push disturbances of up to 130 N.

Abstract

Humanoid Loco-Manipulation

Explore similar work

HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids

Humanoid Rickshaw Pulling: Whole-Body Locomotion under Coupled Wheeled Loads

Dataset-Free Compliant Humanoid Loco-Manipulation with Dynamic Online Posture