SEA-Nav: Efficient Policy Learning for Safe and Agile Quadruped Navigation in Cluttered Environments

Shiyi Chen; Mingye Yang; Haiyi Liu; Haiyan Mao; Jiaqi Zhang; Shuheng He; Debing Zhang; Zihao Qiu

cs.ROMar 10, 2026

SEA-Nav: Efficient Policy Learning for Safe and Agile Quadruped Navigation in Cluttered Environments

Authors: Shiyi Chen, Mingye Yang, Haiyi Liu, Haiyan Mao, Jiaqi Zhang, Shuheng He, Debing Zhang, Zihao Qiu, +1 more

Organizations: Tsinghua University, Beijing, China. · Imperial College London, London, UK.

Abstract

Efficiently learning safe and agile quadruped navigation in densely cluttered environments remains difficult: existing methods often lack safety and agility, or become conservative in complex scenes and require long training schedules. We propose SEA-Nav (Safe, Efficient, and Agile Navigation), a safe reinforcement learning framework for quadruped navigation in cluttered environments. A differentiable control barrier function (CBF) shield constrains the policy to produce safe velocity commands. An adaptive collision-state initialization mechanism increases the probability of learning from safety-critical near-collision experience. An action regularization term further suppresses infeasible commands for physical deployment. The policy converges after about one hour of training on a single RTX 4090 and transfers zero-shot to real-world cluttered scenes.

Figures & tables

Fig. 1: SEA-Nav is trained in about one hour on a single RTX 4090 and deployed zero-shot in a previously unseen maze. The robot successfully escapes using the built-in MPC controller and onboard sparse LiDAR, both unseen during training.

Fig. 2: Overview of the proposed SEA-Nav pipeline. Historical observations are encoded into a latent feature, which is concatenated with the current observation to form the joint state representation. In the Actor, the joint state passes through a shared Backbone into a navigation action head and a safety-gain $\alpha$ head, producing a nominal velocity command and an adaptive gain; the LSE-CBF Shield then solves for a safe velocity command. The Critic takes the joint state representation $x_{t}=[o_{t},z_{t}]$ as input and directly predicts the state value $V(x_{t})$ . The Actor and Critic are optimized jointly with PPO, shield intervention, and kinematic regularization losses.

Fig. 3: SEA-Nav trajectories in three navigation scenarios of increasing difficulty. Each subfigure shows two distinct start-goal trials. SEA-Nav successfully traverses narrow passages and performs timely maneuvering adjustments to avoid entrapment in cluttered environments.

Method	Easy			Medium			Hard
	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	TR $(\%)$ $\downarrow$	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	TR $(\%)$ $\downarrow$	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	TR $(\%)$ $\downarrow$
SEA-Nav (Ours)	100.00 (±0.00)	0.00 (±0.00)	0.00 (±0.00)	97.00 (±0.82)	1.00 (±0.82)	2.00 (±0.00)	90.00 (±1.63)	5.00 (±0.82)	5.00 (±0.82)
SEA-Nav w/o ACSI	100.00 (±0.00)	0.00 (±0.00)	0.00 (±0.00)	93.00 (±0.82)	2.33 (±0.47)	4.67 (±0.47)	83.00 (±2.45)	8.00 (±2.16)	9.00 (±1.63)
SEA-Nav w/o Shield	97.67 (±0.47)	1.33 (±0.47)	1.00 (±0.00)	90.00 (±0.82)	2.67 (±0.94)	7.33 (±1.70)	74.33 (±3.30)	11.67 (±1.70)	14.00 (±2.16)
SEA-Nav w/o $L_{reg}$	96.33 (±0.47)	2.67 (±0.47)	1.00 (±0.00)	82.33 (±2.05)	6.00 (±0.82)	11.67 (±1.25)	57.00 (±4.08)	18.00 (±2.16)	25.00 (±2.16)
ABS [ 7 ]	95.00 (±0.00)	4.67 (±0.47)	0.33 (±0.47)	75.33 (±1.25)	9.33 (±0.47)	15.33 (±1.70)	45.33 (±3.40)	22.00 (±1.41)	32.67 (±2.05)

TABLE I: Different difficulty levels are evaluated; mean and standard deviation are reported over 3 evaluation runs of 100 randomized trials each.

Fig. 4: (Left) Trajectory plot; (right) velocity profiles. SEA-Nav maintains larger obstacle clearance and smoother speed variations. The safety gain $\alpha$ decreases in hazardous regions so the CBF Shield dominates, and increases in safe regions where the nominal navigation command takes the lead.

Fig. 5: Real-world experimental environments. Ten trials are conducted in each environment.

Method	Cluttered Room			Dynamic Obstacle			Obstacle Course			S-Blend Track
Method	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	AS (m/s) $\uparrow$	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	AS (m/s) $\uparrow$	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	AS (m/s) $\uparrow$	SR $(\%)$ $\uparrow$	CR $(\%)$ $\downarrow$	AS (m/s) $\uparrow$
SEA-Nav	100	10	1.6	90	20	1.5	100	10	1.2	100	10	1.4
SEA-Nav-b	100	10	0.9	100	20	0.8	90	10	0.7	100	10	0.5
ABS [ 7 ]	70	50	2.1	60	10	1.7	0	100	0.6	80	20	0.8
OCR [ 23 ]	90	20	1.7	90	10	1.5	30	100	1.0	90	20	1.3
REASAN [ 14 ]	90	20	1.3	90	10	1.3	40	100	0.9	90	30	0.9

TABLE II: Real-world deployment results. AS: average speed (m/s); SEA-Nav-b: built-in MPC controller and onboard sparse LiDAR.

Term	Purpose	Compact expression	Weight
$r_{\text{term}}$	episode termination cost	$\mathbf{1}[\text{terminated}]$	$-100$
$r_{\text{reach}}$	tight goal reaching	$\frac{1}{1+2d^{2}}\,\mathbf{1}[d<0.5]$	$+10$
$r_{\text{velo}}$	move toward goal	$\cos\theta\,v_{x}+\frac{1}{1+2d^{2}}$	$+15$
$r_{\text{clear}}$	clearance-aware motion	$\mathbf{1}[d>1]\,\cos\phi\,v_{x}+\mathbf{1}[d\leq 1]\,\frac{1}{1+2d^{2}}$	$+15$
$r_{\text{stuck}}$	penalize getting stuck	$\mathbf{1}[d>1]\,\mathbf{1}[\Delta p_{\max}<0.1]\,\mathbf{1}[v_{x}>0.0]\,\mathbf{1}[\|\omega_{z}\|<1.0]$	$-5$
$r_{\text{coll}}$	penalize collisions	$\big(1+4(\lVert\mathbf{v}\rVert_{2}^{2}+\omega_{z}^{2})\big)\,\sum_{k}\beta_{k}\,\mathbf{1}[\lVert\mathbf{f}_{k}\rVert>0.1]$	$-4$

TABLE III: Reward terms and weights.

Fig. 6: Comparison of LiDAR point clouds. The robot’s position and orientation are indicated by the gray arrow. (a): The robot’s actual environment. (b): Sparse point cloud from the Unitree Go2 robot’s onboard LiDAR, consisting of approximately 1,000 3D points per frame. (c): Denser point cloud from the RPLIDAR A2, consisting of approximately 2,000 2D points per frame.

Term	Value
Rays delay	$\mathcal{U}(40,80)\,\mathrm{ms}$
Gravity noise	$\mathcal{U}(-0.05,0.05)$
Linear velocity noise	$\mathcal{U}(-0.1,0.1)\,\mathrm{m/s}$
Angular velocity noise	$\mathcal{U}(-0.1,0.1)\,\mathrm{rad/s}$
Friction coefficient factor	$\mathcal{U}(-0.2,1.25)$
Mass perturbation	$\mathcal{U}(-1.5,1.5)\,\mathrm{kg}$

TABLE IV: Domain-randomization parameters.

Term	Value	Term	Value
Buffer length $N_{buf}$	100 steps ( $\approx$ 2 s)	Discount $\gamma$	0.998
Min. history for replay $\Delta_{min}$	20 steps (0.4 s)	GAE $\lambda$	0.95
Rollback $\Delta$ (before clipping)	$\mathcal{U}\{100,\dots,149\}$ steps	Entropy coefficient	0.003
$p_{replay}$	0.8	Desired KL	0.01
$P_{min}$ / $P_{max}$	0.1 / 0.5	Init. noise std	1.5
$d_{up}$ / $d_{down}$	0.5 m / 2.0 m	Episode duration	60 s

TABLE V: Hyperparameters for training.

Explore similar work

Learning Agile Navigation in Crowded Environments for Quadruped Robots
Jul 16, 2026Shuyu Wu, Zeyu Liu, Tianbao Zhang +6Robot Policy Learning Robot Navigation
Shanghai Jiao Tong University, Shanghai 200240, China · State Key Laboratory of High-end Heavy-load Robots, Midea Group, Foshan 528300, China
SafeMind: A Risk-Aware Differentiable Control Framework for Adaptive and Safe Quadruped Locomotion
Apr 10, 2026Zukun Zhang, Kai Shu, Mingqiao MoRobotic Control Robot Safety
The University of Hong Kong, Pokfulam, Hong Kong, China · Alibaba, Beijing, China · University of Chinese Academy of Sciences, No.1 Yanqihu East Rd, Huairou District, Beijing, China
Anticipatory Risk-Guided Reinforcement Learning for Safe Flight Through Dynamic Clutter
Jul 26, 2026Yuchao Mei, Guohao Zhang, Luxia Ai +2Collision Avoidance Risk-Sensitive RL
National Key Laboratory of Science and Technology on Multi-spectral Information Processing, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, China

Jul 16, 2026J/K move · Enter open · S save

Learning Agile Navigation in Crowded Environments for Quadruped Robots

Shuyu Wu, Zeyu Liu, Tianbao Zhang, Fanxing Li+5

Shanghai Jiao Tong University, Shanghai 200240, China · State Key Laboratory of High-end Heavy-load Robots, Midea Group, Foshan 528300, China

Robot Policy Learning Robot Navigation Dynamic Obstacle Avoidance

Open paper AskPDF arXiv

Navigating dynamic and crowded environments presents significant challenges for quadruped robots due to severe sensor occlusion and unpredictable human motion. Existing approaches face a trade-off: model-based methods, such as Velocity Obstacles (VO), theoretically guarantee safety but rely on accurate obstacle motion estimates that often fail in dense crowds, while end-to-end learning methods offer robustness but lack motion prediction capability of obstacles, leading to collisions or conservative behaviors. To solve this, we propose VOP-Nav, a novel navigation system that combines the geometric safety of VO with the agile adaptability of end-to-end learning. Using only local onboard observations, our system avoids explicit obstacle detection and tracking pipelines. The VOP-Net processes multi-frame LiDAR data to implicitly encode dynamic constraints and predict a safe velocity region derived from Velocity Obstacle theory. Importantly, the VO predictions serve a dual role: they are used as input to the navigation policy during inference and as a reward signal during training to encourage safe motion. Evaluations in Isaac Gym demonstrate that VOP-Nav achieves higher success rates than all baselines while balancing locomotion speed and collision avoidance. Real-world deployment on a Unitree Go2 quadruped robot further validates the system's robustness and efficiency in complex indoor and outdoor dynamic environments.

Abstract

Explore similar work

Learning Agile Navigation in Crowded Environments for Quadruped Robots

SafeMind: A Risk-Aware Differentiable Control Framework for Adaptive and Safe Quadruped Locomotion

Anticipatory Risk-Guided Reinforcement Learning for Safe Flight Through Dynamic Clutter