AURASeg: Attention-guided Upsampling with Residual-Assistive Boundary Refinement for Onboard Robot Drivable-Area Segmentation

TL;DR

AURASeg is a novel segmentation framework that enhances boundary accuracy and region detection for robot drivable-area segmentation using attention-guided upsampling and residual boundary refinement, suitable for edge deployment.

Contribution

The paper introduces AURASeg, combining residual boundary refinement and attention-guided upsampling to improve drivable-area segmentation accuracy and boundary precision in resource-constrained environments.

Findings

Outperforms baseline models on multiple datasets

Achieves higher boundary accuracy metrics

Demonstrates real-time deployment on Jetson Nano

Abstract

Free space ground segmentation is essential to navigate autonomous robots, recognize drivable zones, and traverse efficiently. Fine-grained features remain challenging for existing segmentation models, particularly for robots in indoor, outdoor and road-scene environments. These difficulties arise from ineffective multi-scale processing, sub-optimal boundary refinement, and limited feature representation. To address this, we propose Attention-guided Upsampling with Residual-Assistive Boundary Refinement (AURASeg), a ground-plane drivable area segmentation framework designed to improve boundary precision while preserving strong region accuracy under edge-deployment constraints. Built on ResNet backbone, we propose (i) a Residual Boundary Refinement Module (RBRM) that enhances edge delineation through boundary-assistive feature refinement, and (ii) Attention Progressive Upsampling Decoder (APUD) blocks that fuse multi-level features using residual fusion of attention modules; additionally, we integrate (iii) a lightweight ASPPLite module to capture multi-scale context with minimal overhead. Extensive experiments on CARL-D, the Ground Mobile Robot Perception (GMRPD) dataset, and a custom Gazebo indoor dataset show that AURASeg consistently outperforms strong baselines, with notable gains in boundary metrics. Finally, we demonstrate on-device deployment on a Jetson Nano powered Kobuki TurtleBot, validating practical edge-inference feasibility. Code is omitted for anonymity and will be released upon acceptance.