HCSG combines geometric forecasting of human pose and trajectory with VLM-generated semantic descriptions of intentions, fused into a topological map with a social distance loss, yielding 14% higher success rate and 34% lower collision rate on the HA-VLNCE benchmark.
Social-llava: Enhancing robot navigation through human-language reasoning in social spaces
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
fields
cs.RO 5verdicts
UNVERDICTED 5representative citing papers
iCrowdNav encodes egocentric visual observations with occupancy features and human pose intentions to improve DRL policies for crowd navigation, showing better performance than baselines in experiments and real-world tests.
SALSA aligns social features and adds future-risk signals in VLA models to cut near-collisions by 86.4% and raise social accuracy from 53% to 93% on SCAND and real robots.
AutoSpatial improves VLM spatial reasoning for social navigation by combining minimal manual supervision with auto-labeled VQA pairs and hierarchical training, showing gains up to 20.5% in action prediction over baselines.
Multimodal explainability module using vision-language models and heat maps enables robots to generate natural-language summaries of navigation observations, with n=30 user studies showing majority preference for real-time explanations and improved trust.
citing papers explorer
-
HCSG: Human-Centric Semantic-Geometric Reasoning for Vision-Language Navigation
HCSG combines geometric forecasting of human pose and trajectory with VLM-generated semantic descriptions of intentions, fused into a topological map with a social distance loss, yielding 14% higher success rate and 34% lower collision rate on the HA-VLNCE benchmark.
-
Learning Robot Visual Navigation in Crowds via Intention-Aware Scene Representations
iCrowdNav encodes egocentric visual observations with occupancy features and human pose intentions to improve DRL policies for crowd navigation, showing better performance than baselines in experiments and real-world tests.
-
Act on What You See: Unlocking Safe Social Navigation in Vision-Language-Action Models
SALSA aligns social features and adds future-risk signals in VLA models to cut near-collisions by 86.4% and raise social accuracy from 53% to 93% on SCAND and real robots.
-
AutoSpatial: Visual-Language Reasoning for Social Robot Navigation through Efficient Spatial Reasoning Learning
AutoSpatial improves VLM spatial reasoning for social navigation by combining minimal manual supervision with auto-labeled VQA pairs and hierarchical training, showing gains up to 20.5% in action prediction over baselines.
-
Trust Through Transparency: Explainable Social Navigation for Autonomous Mobile Robots via Vision-Language Models
Multimodal explainability module using vision-language models and heat maps enables robots to generate natural-language summaries of navigation observations, with n=30 user studies showing majority preference for real-time explanations and improved trust.