EffiNav: Fusing Depth and Vision-Language for Efficient Object Goal Navigation

arXiv:2606.18634 · cs.RO, cs.AI · Submitted 2026-06-17 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: We're looking at the paper: "EffiNav: Fusing Depth and Vision-Language for Efficient Object Goal Navigation".

Dev: Here's the gist of the paper. To locate a target object while exploring an unknown environment is a fundamental capability for autonomous agents, with applications ranging from search-and-rescue to field robots.

Rosa: The paper proposes EffiNav, a training-free framework for efficient object goal navigation that harnesses the reasoning capabilities of vanilla VLMs to provide intelligent, vision-driven guidance for navigation control.

Dev: Problem Statement and Motivation ObjNav is a one-stage navigation task which requires robot to directly go to the target object in an unknown environment.

Rosa: The key to efficient navigation lies in deciding where to explore next.

Title and authors: Rosa: So we’re diving into the specifics of this paper, "EffiNav: Fusing Depth and Vision-Language for Efficient Object Goal Navigation." We’ll talk about what they are actually doing in their framework.

Dev: The paper outlines this whole system as a training-free method for Object Goal Navigation, which means no new training data is required to get it working. They focus on making navigation efficient while maintaining success rates.

Taro: What I’m curious about is how they structure that efficiency part since you mentioned the goal is to find the object without just wandering around aimlessly. Is there a clear mechanism for deciding where to go?

Rosa: They have this intelligent selection-and-verification scheme, which they call Module A, and it starts with depth thresholds, like looking for areas further than five meters away. These areas become candidate exploration options that the VLM then looks at using prompts.

Dev: And those prompts are essentially Vision Question Answering tasks, like asking the VLM "To find the bed, which of the given areas is the best to reach?". They use RGB information from those candidate regions to give the reasoning core context.

The paper's summary: Rosa: So that brings us to what they actually propose in EffiNav. It’s a depth-aware framework that uses an open-source VLM as its decision-making core, and they’ve built in modules to stop the robot from re-exploring old areas.

Dev: They have this whole flow: first, you get area masks from depth thresholds, then the VLM reasons about those candidates using a Vision Question Answering task. Then there's another step where they perform global consistency checking by projecting those selected regions onto a top-down map and feeding that back into a second VQA query.

Taro: That top-down check sounds smart because it makes the decision not just local to what the robot sees right now, but also consistent with the overall layout of the environment. It's like checking if a path looks good from above, not just from your current spot.

Rosa: Precisely. Then they add this history-aware pruning mechanism to remove candidate regions that are too close to where the robot has already explored. It’s designed to prevent those wasteful back-and-forth motions we see in older navigation strategies.

The paper's improvements: Dev: Now let’s talk about how they improve things compared to what came before. They introduce several modules that build on the core idea of using VLM reasoning to drive control.

Rosa: Module A, as we talked about, is the depth-aware intelligent selection mechanism where RGB context from candidate regions informs the VLM's decision. That makes the initial exploration selection much richer than just using raw depth data alone.

Taro: I’m interested in Module B, which is the global consistency verification. They project those selected egocentric regions onto a top-down map to verify spatial reasonableness before committing to an area, and then they use that as input for a second VQA query. That seems like a strong way to check if the choice makes sense globally.

Dev: And they also have this history-aware pruning mechanism. It’s designed specifically to prevent those redundant explorations by removing candidate regions within a predefined distance of the already explored area, ensuring we don't waste time going back where we just came from.

Conclusion: Rosa: So to wrap up, EffiNav is essentially this training-free framework that uses depth and VLM reasoning together for efficient Object Goal Navigation. It aims to balance success rate with trajectory efficiency using a new metric called Normalized Efficiency on Successes, or EoS.

Dev: The results are pretty solid; they show a mean EoS of sixty-six point thirty-three on the HMthree dee benchmark and a median of sixty-five point thirty-five on OVON. They claim this framework outperforms almost all the baselines in both success weighted by path length and achieves the highest EoS on both those benchmarks.

Taro: What stands out for me is that they show this consistency across different scenes, even when things change, which suggests a level of robustness we don't always see in these types of systems. I’m also interested in the failure analysis they did, where they found that navigation-related issues like Constraint Map and Timeout were dominant on HMthree dee.

Rosa: It seems the biggest takeaway is that you can get stable, balanced performance without spending time training a model or collecting huge amounts of specific data for every new environment. EffiNav shows how leveraging existing VLM reasoning power can lead to efficient navigation in real settings.

Zecheng Yin, Benedict Jun Ma

Systems Hub of Intelligence Transportation · HKUST(GZ)

cs.RO, cs.AI

Submitted: 2026-06-17

Updated: 2026-11-09

Comments: I wrongly issued the institute of this paper, and I will resubmit this after several month with right institute name.

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 80/100

The gist: To locate a target object while exploring an unknown environment is a fundamental capability for autonomous agents, with applications ranging from search-and-rescue to field robots.

Key concepts

EffiNav
A training-free framework designed for efficient object goal navigation. It uses the reasoning power of vanilla VLMs combined with depth perception to guide a robot directly to a target object in an unknown environment, optimizing the exploration path.
Vision Question Answering (VQA)
A task where the VLM is asked questions about visual input, such as 'Which area is best to reach the bed?'. This allows the system to use natural language reasoning based on what it sees in an image and its understanding of spatial relationships.
Normalized Efficiency on Successes (EoS)
A new metric measuring how efficiently an agent reaches a goal when it succeeds. It is calculated as the ratio of path length to success rate, providing a single score between 0 and 1 that quantifies the trajectory's efficiency.
History-aware Pruning
A mechanism that stops the robot from re-exploring areas it has already visited within a certain distance. This prevents redundant back-and-forth motion by intelligently ignoring candidate regions close to previously explored locations.

Terminology

Summary

To locate a target object while exploring an unknown environment is a fundamental capability for autonomous agents, with applications ranging from search-and-rescue to field robots. p. 0

Where the paper says it

EffiNav: Fusing Depth and Vision-Language for Efficient Object Goal Navigation Zecheng Yin, Benedict Jun Ma Systems Hub of Intelligence Transportation HKUST(GZ) Guangzhou, Guangdong Abstract—To locate a target object while exploring the unknown environment is a fundamental capability for autonomous agents, with applications ranging from search-and-rescue to field robots. A simplified version of such task is Object Goal Navigation (ObjNav).

The paper proposes EffiNav, a training-free framework for efficient object goal navigation that harnesses the reasoning capabilities of vanilla VLMs to provide intelligent, vision-driven guidance for navigation control. p. 8
Where the paper says it

CONCLUSION In this paper, we propose EffiNav, a training-free framework for efficient object-goal navigation (ObjNav) that delivers stable, balanced performance and exhibits extensibility to versatile navigation tasks. EffiNav harnesses the reasoning capabilities of vanilla VLM to provide intelligent, vision-driven guidance for navigation control, to achieve efficient ObjNav.

Problem Statement and Motivation

ObjNav is a one-stage navigation task which requires robot to directly go to the target object in an unknown environment. The key to efficient navigation lies in deciding where to explore next. p. 0

Where the paper says it

In ObjNav, successful arrival at the target object provides a basic measure of performance; however, the efficiency of the navigation trajectory is equally important, as it indicates how intelligently the agent explores and how much time remains for subsequent tasks. In unknown environments, the key to efficient navigation lies in deciding where to explore next.

Prior works suffer from generalization and efficiency issues, which can lead to excessive exploration of already-visited areas or redundant back-and-forth motion. p. 0
Where the paper says it

In unknown environments, the key to efficient navigation lies in deciding where to explore next. While many prior works aim to address this core challenge and achieved promising performance in certain settings, recent training-based models and non-training frameworks still suffer from generalization and efficiency issues respectively, which in the worst cases can lead to excessive exploration of already-visited areas or redundant back-and-forth motion.

The paper distinguishes ObjNav from Vision-andLanguage Navigation in Continuous Environments (VLNCE) because ObjNav provides no environmental prior at initialization. p. 0
Where the paper says it

It is important to distinguish ObjNav from Vision-andLanguage Navigation in Continuous Environments (VLNCE) [10]. In VLN-CE, the agent receives complex, temporally or spatially extended natural language instructions that operates with prior information [15, 11, 5].

Framework Architecture

EffiNav is a depth-aware intelligent nontraining framework for ObjNav, utilizing reasoning ability of opensource VLMs to perceiving the surroundings and depth to drive the robots and incorporates modules to prevent history re-explorations. The perception and intelligent guidance core involves Module a (Ego Region Selecting), which uses depth thresholds to generate potential area masks, followed by a Vision Question Answering (VQA) task with prompts such as “To find the bed, which of the given areas is the best to reach? p. 2

Where the paper says it

Following prior work [25, 6, 8, 26], this process is formulated as a Vision Question Answering (VQA) task, with prompts such as “To find the bed, which of the given areas is the best to reach? This module ensures that the decision core receives rich and immediate feedback about the surrounding environment and available exploration candidates.

”. Module b (Global-wise Checking) performs global consistency verification by projecting selected regions onto a top-down map and feeding this back as visual input to a second VQA query. p. 2
Where the paper says it

Module b (Global-wise Checking) performs global consistency verification. After the VLM selects the best candidate region in the egocentric view, this region is projected onto the gradually constructed top-down map and highlighted with a distinct color.

A history-aware pruning mechanism is incorporated to remove candidate regions within a predefined distance of the explored area. p. 2
Where the paper says it

If the decision core rejects the choice, the framework discards this candidate and returns to the previous selection step (Module a) to choose second best alternative. To prevent redundant exploration, we incorporate a history-aware pruning mechanism: any candidate region that lies within a predefined distance of the explored area (the gray area in Figure 1) boundary is considered valid; regions beyond this margin are deemed already sufficiently explored and are removed from the candidate pool.

Performance Evaluation and Metrics

ObjNav is evaluated using two complementary metrics: Success Rate (SR), the fraction of episodes in which the target object is successfully reached, and Success weighted by Path Length (SPL), which penalizes inefficient trajectories by normalizing success with respect to path length. The paper introduces a new metric, Normalized Efficiency on Successes (EoS), formulated in Equation 2, which quantifies how efficiently an agent reaches the goal when it succeeds. p. 1

Where the paper says it

To better capture navigation efficiency among successful trials, we introduce a new metric Normalized Efficiency on Successes (EoS), formulated in Equation 2. EoS quantifies how efficiently an agent reaches the goal when it succeeds, providing a finergrained assessment beyond SR and SPL.

EoS is computed as the ratio of SPL to SR, and it ranges from 0 to 1. p. 4
Where the paper says it

We introduce a new metric, Normalized Efficiency on Successes (EoS), which purely measures navigation efficiency conditioned on successful episodes, as defined in Eq. EoS is computed as the ratio of SPL to SR.

Experimental Results and Analysis

EffiNav demonstrates consistent performance across different scenes, with a mean EoS of 66.33 on HM3D and a median of 65. p. 4

Where the paper says it

HM3D has a higher mean geodesic distance (7. 60 m) and median (6.

35 on OVON. p. 0
Where the paper says it

While many prior works aim to address this core challenge and achieved promising performance in certain settings, recent training-based models and non-training frameworks still suffer from generalization and efficiency issues respectively, which in the worst cases can lead to excessive exploration of already-visited areas or redundant back-and-forth motion. We evaluate EffiNav on two widely used simulation benchmarks Habitat Matterport 3D (HM3D) and Open-Vocabulary Object goal Navigation (OVON), and further validate its effectiveness on physical robots in realworld settings.

The framework outperforms nearly all baselines in SPL and achieves the highest EoS on both HM3D and OVON. p. 5
Where the paper says it

Our framework performs efficient navigation. As shown in Table V, EffiNav outperforms nearly all baselines in SPL and achieves the highest EoS on both HM3D and OVON.

The analysis shows that EffiNav maintains consistent EoS despite environmental variations, highlighting its robustness. p. 5
Where the paper says it

Notably, even though all scenes are decently reconstructed, the absence constructions in house layouts vary significantly, leading to substantial performance fluctuations [21], as reflected in unstable SR and SPL across scenes, especially in depth-aware methods like ours. Despite of that, EffiNav maintains consistent EoS despite environmental variations, highlighting its robustness.

Conclusion and Limitations

EffiNav is a training-free framework for efficient ObjNav that delivers stable, balanced performance and exhibits extensibility to versatile navigation tasks. A limitation of the approach is that it relies on 2D images for both egocentric observations and top-down views, which significantly sacrifices spatial information. p. 7

Where the paper says it

First, as our approach relies on a VLM for control, it requires an accurate and informative representation of the current environment. However, both the egocentric observations and the top-down views used in our pipeline are 2D images, which significantly sacrifice spatial information.

Furthermore, finalization of navigation relies on 2D object detection which is sensitive to viewpoint variations. p. 7
Where the paper says it

In our experiments, we attempted to use estimated depth from RGB images using the method of [12], but the results were unsatisfactory. Third, finalization of navigation relies on 2D object detection which is sensitive to viewpoint variations.

The framework's performance is critically dependent on accurate depth estimates, as errors in depth measurement directly affect the spatial projection of VLM-generated guidance.

Extensibility and Real-World Application

The paper extends EffiNav to a memory-augmented ObjNav task on GOAT-BENCH, demonstrating its adaptability beyond standard ObjNav settings. In real-world validation, the robot was able to locate the target object without fully exploring the environment or reconstructing the entire house. p. 6

Where the paper says it

Trajectories on the same goal “refrigerator cabinet” with different starting position in the same room, ending up with the same observation, demonstrating the consistent exploration strategy of our framework. Case study As shown in Figure 8 and Figure 9, our robot was able to locate the target object without fully exploring the environment or reconstructing the entire house.

The efficiency of ObjNav can drop considerably without intelligent perception and ego-centric and global wise guidance in real-world scenarios. p. 7
Where the paper says it

Even with the same detection ability, without the intelligent perception and ego-centric and global wise guidance, the efficiency of the ObjNav can drop considerably. TABLE VII REAL-WORLD ABLATION ON GOAL “TEDDY BEAR ON SOFA”.

Failure Analysis

Failure statistics across two benchmarks show that main causes are Constraint Map, Semantic Understanding, Category Mismatch, and Others. In HM3D, the dominant cause is navigation-related (primarily Constraint Map and Timeout), reflecting its emphasis on robust exploration and mobility. p. 6

Where the paper says it

Notably, the distribution of failure modes differs between the two datasets. In HM3D, the dominant cause is navigationrelated (primarily Constraint Map and Timeout), reflecting its emphasis on robust exploration and mobility.

In contrast, OVON exhibits a higher proportion attributed to Semantic Understanding and Category Mismatch. p. 6
Where the paper says it

In HM3D, the dominant cause is navigationrelated (primarily Constraint Map and Timeout), reflecting its emphasis on robust exploration and mobility. In contrast, OVON exhibits a higher proportion attributed to Semantic Understanding and Category Mismatch, emphasizing its focus on object grounding under diverse goals.

The distribution of failure modes differs between the two datasets, emphasizing its focus on object grounding under diverse goals. p. 6
Where the paper says it

In HM3D, the dominant cause is navigationrelated (primarily Constraint Map and Timeout), reflecting its emphasis on robust exploration and mobility. In contrast, OVON exhibits a higher proportion attributed to Semantic Understanding and Category Mismatch, emphasizing its focus on object grounding under diverse goals.

Summary of Findings

EffiNav outperforms recent state-of-the-art foundation models and frameworks in both navigation capability and target generalization. The core design principle is prioritizing navigation efficiency in a training-free framework, leveraging depth-aware perception and joint confirmation from egocentric observations and top-down visual-language reasoning. p. 5

Where the paper says it

This stems from our core design principle: prioritizing navigation efficiency in a training-free framework. The depthaware perception and the joint confirmation from egocentric observations and top-down visual-language reasoning together contribute efficient decision-making in ObjNav.

This approach rivals trained foundation models without any cost of training or data collection. p. 5
Where the paper says it

Our framework outperforms non-training baselines and rivals trained foundation models. Remarkably, EffiNav remains highly competitive against large-scale trained models yet without any cost of training or data collection.

The framework achieves best performance in both two aspects when compared with different baselines. p. 6
Where the paper says it

As illustrated in the Table VI, our framework achieved best performance in both two aspects. Our SPL is far surpass baselines, demonstrating again our framework performs a steady efficient navigation ability.

Memory-Augmented Extension

In the Memory ObjNav task on GOAT-BENCH, our framework achieved best performance in both two aspects. Our SPL is far surpass baselines, demonstrating again our framework performs a steady efficient navigation ability. p. 6

Where the paper says it

As illustrated in the Table VI, our framework achieved best performance in both two aspects. Our SPL is far surpass baselines, demonstrating again our framework performs a steady efficient navigation ability.

This reveals that with easy extension, our framework can be applied for many other tasks with potential considerable efficiency improvement. p. 6
Where the paper says it

Our SPL is far surpass baselines, demonstrating again our framework performs a steady efficient navigation ability. This reveals that with easy extension, our framework can be applied for many other tasks with potential considerable efficiency improvement.

Real-World Validation Results

In real-world validation on a Unitree Go2 robot in an indoor office, EffiNav showed an obvious advantage in the EoS and SPL compared to BFS nearest-frontier exploration. Even with the same detection ability, without the intelligent perception and ego-centric and global wise guidance, the efficiency of ObjNav can drop considerably. p. 7

Where the paper says it

As we see both of the methods have fairly steady SR, but they differ very much on the EoS and SPL, where EffiNav has obvious advantage. Even with the same detection ability, without the intelligent perception and ego-centric and global wise guidance, the efficiency of the ObjNav can drop considerably.

The top-down map of the real-world environment shows that obstacle shapes are incomplete or missing due to sensor inaccuracies and limited sensor range. p. 7
Where the paper says it

This gap tends to widen, especially in larger rooms. In the top-down map of the real-world environment, obstacle shapes are incomplete or missing due to sensor inaccuracies and limited sensor range.

Conclusion

In this paper, we propose EffiNav, a training-free framework for efficient object goal navigation (ObjNav) that delivers stable, balanced performance and exhibits extensibility to versatile navigation tasks. EffiNav harnesses the reasoning capabilities of vanilla VLM to provide intelligent, vision-driven guidance for navigation control, to achieve efficient ObjNav. p. 8

Where the paper says it

CONCLUSION In this paper, we propose EffiNav, a training-free framework for efficient object-goal navigation (ObjNav) that delivers stable, balanced performance and exhibits extensibility to versatile navigation tasks. EffiNav harnesses the reasoning capabilities of vanilla VLM to provide intelligent, vision-driven guidance for navigation control, to achieve efficient ObjNav.

This work is not an endpoint. p. 8
Where the paper says it

Furthermore, our extension to memory-augmented navigation highlights the framework’s adaptability and potential for broader applications. This work is not an endpoint.

The potential of our framework, which built upon their evolving reasoning power, will only grow stronger. p. 8
Where the paper says it

This work is not an endpoint. As VLMs continue to advance rapidly, the potential of our framework, which built upon their evolving reasoning power, will only grow stronger.

Acknowledgments

VQA text prompt, simulation and real-world navigation videos, Efficiency on Successes (EoS) Derivation are in supplemental materials. The detection service of Qwen-VL 2. p. 7

Where the paper says it

We extract the point cloud, depth, rgb information from the sensor via pyk4a using a self-made server in unit16 format. Both methods use the same detection and segmentation service provided by [22] and [32].

5 model is a stable vision-language inference service provided by the official Qwen team. p. 9
Where the paper says it

IEEE Robotics and Automation Letters, 10(7):6576–6583, 2025. [22] Qwen Team. p. 9

Where the paper says it

IEEE Robotics and Automation Letters, 10(7):6576–6583, 2025. [22] Qwen Team.

The detection service has demonstrated high reliability, introducing only minimal performance variation in our long-term experiments conducted over more than five months.

References

[1] Peter Anderson, Angel Chang, Devendra Singh Chaplot, Alexey Dosovitskiy, Saurabh Gupta, Vladlen Koltun, Jana Kosecka, Jitendra Malik, Roozbeh Mottaghi, Manolis Savva, and Amir R. Zamir. On evaluation of embodied navigation agents. p. 8

Where the paper says it

REFERENCES [1] Peter Anderson, Angel Chang, Devendra Singh Chaplot, Alexey Dosovitskiy, Saurabh Gupta, Vladlen Koltun, Jana Kosecka, Jitendra Malik, Roozbeh Mottaghi, Manolis Savva, and Amir R. On evaluation of embodied navigation agents, 2018.

p. 8
Where the paper says it

REFERENCES [1] Peter Anderson, Angel Chang, Devendra Singh Chaplot, Alexey Dosovitskiy, Saurabh Gupta, Vladlen Koltun, Jana Kosecka, Jitendra Malik, Roozbeh Mottaghi, Manolis Savva, and Amir R. On evaluation of embodied navigation agents, 2018.

p. 8
Where the paper says it

REFERENCES [1] Peter Anderson, Angel Chang, Devendra Singh Chaplot, Alexey Dosovitskiy, Saurabh Gupta, Vladlen Koltun, Jana Kosecka, Jitendra Malik, Roozbeh Mottaghi, Manolis Savva, and Amir R. On evaluation of embodied navigation agents, 2018.

[2] Wenzhe Cai, Jiaqi Peng, Yuqiang Yang, Yujian Zhang, Meng Wei, Hanqing Wang, Yilun Chen, Tai Wang, and Jiangmiao Pang. p. 8

Where the paper says it

org/abs/1807. [2] Wenzhe Cai, Jiaqi Peng, Yuqiang Yang, Yujian Zhang, Meng Wei, Hanqing Wang, Yilun Chen, Tai Wang, and Jiangmiao Pang.

Navdp: Learning sim-to-real navigation diffusion policy with privileged information guidance. p. 8
Where the paper says it

[2] Wenzhe Cai, Jiaqi Peng, Yuqiang Yang, Yujian Zhang, Meng Wei, Hanqing Wang, Yilun Chen, Tai Wang, and Jiangmiao Pang. Navdp: Learning sim-to-real navigation diffusion policy with privileged information guidance, 2025.

[3] Yihan Cao, Jiazhao Zhang, Zhinan Yu, Shuzhen Liu, Zheng Qin, Qin Zou, Bo Du, and Kai Xu. p. 8

Where the paper says it

Navdp: Learning sim-to-real navigation diffusion policy with privileged information guidance, 2025. [3] Yihan Cao, Jiazhao Zhang, Zhinan Yu, Shuzhen Liu, Zheng Qin, Qin Zou, Bo Du, and Kai Xu.

Cognav: Cognitive process modeling for object goal navigation with llms. p. 8
Where the paper says it

[3] Yihan Cao, Jiazhao Zhang, Zhinan Yu, Shuzhen Liu, Zheng Qin, Qin Zou, Bo Du, and Kai Xu. Cognav: Cognitive process modeling for object goal navigation with llms.

[4] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng, and Yinda Zhang. p. 8

Where the paper says it

In International Conference on Computer Vision (ICCV), 2025. [4] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng, and Yinda Zhang.

Matterport3D: Learning from RGB-D data in indoor environments. p. 8
Where the paper says it

[4] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng, and Yinda Zhang. Matterport3D: Learning from RGB-D data in indoor environments.

p. 8
Where the paper says it

[4] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng, and Yinda Zhang. Matterport3D: Learning from RGB-D data in indoor environments.

[5] An-Chieh Cheng, Yandong Ji, Zhaojing Yang, Xueyan Zou, Jan Kautz, Erdem Biyik, Hongxu Yin, Sifei Liu, and Xiaolong Wang. p. 8

Where the paper says it

International Conference on 3D Vision (3DV), 2017. [5] An-Chieh Cheng, Yandong Ji, Zhaojing Yang, Xueyan Zou, Jan Kautz, Erdem Biyik, Hongxu Yin, Sifei Liu, and Xiaolong Wang.

Navila: Legged robot vision-language action model for navigation. p. 8
Where the paper says it

[5] An-Chieh Cheng, Yandong Ji, Zhaojing Yang, Xueyan Zou, Jan Kautz, Erdem Biyik, Hongxu Yin, Sifei Liu, and Xiaolong Wang. Navila: Legged robot vision-languageaction model for navigation.

[6] Dylan Goetting, Himanshu Gaurav Singh, and Antonio Loquercio. p. 8

Where the paper says it

In RSS, 2025. [6] Dylan Goetting, Himanshu Gaurav Singh, and Antonio Loquercio.

End-to-end navigation with vlms: Transforming spatial reasoning into question-answering. p. 8
Where the paper says it

[6] Dylan Goetting, Himanshu Gaurav Singh, and Antonio Loquercio. End-to-end navigation with vlms: Transforming spatial reasoning into question-answering.

[7] Yiming Ji, Kaijie Yun, Yang Liu, Zhengpu Wang, Boyu Ma, Zongwu Xie, and Hong Liu. p. 8

Where the paper says it

In Workshop on Language and Robot Learning: Language as an Interface, 2024. [7] Yiming Ji, Kaijie Yun, Yang Liu, Zhengpu Wang, Boyu Ma, Zongwu Xie, and Hong Liu.

Diffusion as reasoning: Enhancing object navigation via diffusion model conditioned on llm-based object-room knowledge. p. 8
Where the paper says it

[7] Yiming Ji, Kaijie Yun, Yang Liu, Zhengpu Wang, Boyu Ma, Zongwu Xie, and Hong Liu. Diffusion as reasoning: Enhancing object navigation via diffusion model conditioned on llm-based object-room knowledge, 2025.

[8] Zihe Ji, Huangxuan Lin, and Yue Gao. p. 8

Where the paper says it

org/abs/2410. [8] Zihe Ji, Huangxuan Lin, and Yue Gao.

Dynavlm: Zeroshot vision-language navigation system with dynamic viewpoints and self-refining graph memory. p. 8
Where the paper says it

[8] Zihe Ji, Huangxuan Lin, and Yue Gao. Dynavlm: Zeroshot vision-language navigation system with dynamic viewpoints and self-refining graph memory, 2025.

[9] Mukul Khanna, Ram Ramrakhya, Gunjan Chhablani, Sriram Yenamandra, Theophile Gervet, Matthew Chang, Zsolt Kira, Devendra Singh Chaplot, Dhruv Batra, and Roozbeh Mottaghi. p. 8

Where the paper says it

org/abs/2506. [9] Mukul Khanna*, Ram Ramrakhya*, Gunjan Chhablani, Sriram Yenamandra, Theophile Gervet, Matthew Chang, Zsolt Kira, Devendra Singh Chaplot, Dhruv Batra, and Roozbeh Mottaghi.

Goat-bench: A benchmark for multimodal lifelong navigation. p. 8
Where the paper says it

[9] Mukul Khanna*, Ram Ramrakhya*, Gunjan Chhablani, Sriram Yenamandra, Theophile Gervet, Matthew Chang, Zsolt Kira, Devendra Singh Chaplot, Dhruv Batra, and Roozbeh Mottaghi. Goat-bench: A benchmark for multimodal lifelong navigation, 2024.

[10] Jacob Krantz, Erik Wijmans, Arjun Majumdar, Dhruv Batra, and Stefan Lee. p. 8

Where the paper says it

Goat-bench: A benchmark for multimodal lifelong navigation, 2024. [10] Jacob Krantz, Erik Wijmans, Arjun Majumdar, Dhruv Batra, and Stefan Lee.

Beyond the nav-graph: Visionand-language navigation in continuous environments. p. 8
Where the paper says it

[10] Jacob Krantz, Erik Wijmans, Arjun Majumdar, Dhruv Batra, and Stefan Lee. Beyond the nav-graph: Visionand-language navigation in continuous environments.

[11] Zerui Li, Gengze Zhou, Haodong Hong, Yanyan Shao, Wenqi Lyu, Yanyuan Qiao, and Qi Wu. p. 8

Where the paper says it

1007/978-3-030-58604-1 7. [11] Zerui Li, Gengze Zhou, Haodong Hong, Yanyan Shao, Wenqi Lyu, Yanyuan Qiao, and Qi Wu.

Ground-level viewpoint vision-and-language navigation in continuous environments. p. 8
Where the paper says it

[11] Zerui Li, Gengze Zhou, Haodong Hong, Yanyan Shao, Wenqi Lyu, Yanyuan Qiao, and Qi Wu. Ground-level viewpoint vision-and-language navigation in continuous environments.

[12] Haotong Lin, Sili Chen, Jun Hao Liew, Donny Y. p. 8

Where the paper says it

1109/ICRA55743. [12] Haotong Lin, Sili Chen, Jun Hao Liew, Donny Y.

Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang. p. 8
Where the paper says it

[12] Haotong Lin, Sili Chen, Jun Hao Liew, Donny Y. Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang.

Depth anything 3: Recovering the visual space from any views. p. 8
Where the paper says it

Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang. Depth anything 3: Recovering the visual space from any views.

[13] Soroush Nasiriany, Fei Xia, Wenhao Yu, Ted Xiao, Jacky Liang, Ishita Dasgupta, Annie Xie, Danny Driess, Ayzaan Wahid, Zhuo Xu, Quan Vuong, Tingnan Zhang, Tsang-Wei Edward Lee, Kuang-Huei Lee. p. 8

Where the paper says it

10647, 2025. [13] Soroush Nasiriany, Fei Xia, Wenhao Yu, Ted Xiao, Jacky Liang, Ishita Dasgupta, Annie Xie, Danny Driess, Ayzaan Wahid, Zhuo Xu, Quan Vuong, Tingnan Zhang, Tsang-Wei Edward Lee, Kuang-Huei Lee, Peng Xu, Sean Kirmani, Yuke Zhu, Andy Zeng, Karol Hausman, Nicolas Heess, Chelsea Finn, Sergey Levine, and Brian Ichter.

PIVOT: iterative visual prompting elicits actionable knowledge for vlms. p. 8
Where the paper says it

[13] Soroush Nasiriany, Fei Xia, Wenhao Yu, Ted Xiao, Jacky Liang, Ishita Dasgupta, Annie Xie, Danny Driess, Ayzaan Wahid, Zhuo Xu, Quan Vuong, Tingnan Zhang, Tsang-Wei Edward Lee, Kuang-Huei Lee, Peng Xu, Sean Kirmani, Yuke Zhu, Andy Zeng, Karol Hausman, Nicolas Heess, Chelsea Finn, Sergey Levine, and Brian Ichter. PIVOT: iterative visual prompting elicits actionable knowledge for vlms.

[14] Xavi Puig, Eric Undersander, Andrew Szot, Mikael Dallaire Cote, Ruslan Partsey, Jimmy Yang, Ruta Desai, Alexander William Clegg, Michal Hlavac, Tiffany Min. p. 8

Where the paper says it

id=051jaf8MQy. [14] Xavi Puig, Eric Undersander, Andrew Szot, Mikael Dallaire Cote, Ruslan Partsey, Jimmy Yang, Ruta Desai, Alexander William Clegg, Michal Hlavac, Tiffany Min, Theo Gervet, Vladim´ır Vondrus, Vincent-Pierre Berges, John Turner, Oleksandr Maksymets, Zsolt Kira, Mrinal Kalakrishnan, Jitendra Malik, Devendra Singh Chaplot, Unnat Jain, Dhruv Batra, Akshara Rai, and Roozbeh Mottaghi.

Habitat 3.0: A co-habitat for humans and robots. p. 8
Where the paper says it

[14] Xavi Puig, Eric Undersander, Andrew Szot, Mikael Dallaire Cote, Ruslan Partsey, Jimmy Yang, Ruta Desai, Alexander William Clegg, Michal Hlavac, Tiffany Min, Theo Gervet, Vladim´ır Vondrus, Vincent-Pierre Berges, John Turner, Oleksandr Maksymets, Zsolt Kira, Mrinal Kalakrishnan, Jitendra Malik, Devendra Singh Chaplot, Unnat Jain, Dhruv Batra, Akshara Rai, and Roozbeh Mottaghi. 0: A co-habitat for humans, avatars and robots, 2023.

[15] Yanyuan Qiao, Wenqi Lyu, Hui Wang, Zixu Wang, Zerui Li, Yuan Zhang, Mingkui Tan, and Qi Wu. p. 8

Where the paper says it

0: A co-habitat for humans, avatars and robots, 2023. [15] Yanyuan Qiao, Wenqi Lyu, Hui Wang, Zixu Wang, Zerui Li, Yuan Zhang, Mingkui Tan, and Qi Wu.

Opennav: Exploring zero-shot vision-and-language navigation in continuous environment with open-source llms. p. 8
Where the paper says it

[15] Yanyuan Qiao, Wenqi Lyu, Hui Wang, Zixu Wang, Zerui Li, Yuan Zhang, Mingkui Tan, and Qi Wu. Opennav: Exploring zero-shot vision-and-language navigation in continuous environment with open-source llms.

[16] Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin. p. 9

Where the paper says it

Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2025. [16] Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever.

Learning transferable visual models from natural language supervision. p. 9
Where the paper says it

[16] Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision, 2021.

[17] Santhosh K. p. 9

Where the paper says it

org/abs/2103. [17] Santhosh K.

p. 9
Where the paper says it

org/abs/2103. [17] Santhosh K.

Ramakrishnan, Aaron Gokaslan, Erik Wijmans, Oleksandr Maksymets, Alex Clegg, John Turner, Eric Undersander, Wojciech Galuba. p. 9
Where the paper says it

[17] Santhosh K. Ramakrishnan, Aaron Gokaslan, Erik Wijmans, Oleksandr Maksymets, Alex Clegg, John Turner, Eric Undersander, Wojciech Galuba, Andrew Westbury, Angel X.

p. 9
Where the paper says it

[17] Santhosh K. Ramakrishnan, Aaron Gokaslan, Erik Wijmans, Oleksandr Maksymets, Alex Clegg, John Turner, Eric Undersander, Wojciech Galuba, Andrew Westbury, Angel X.

On evaluation of embodied navigation agents.

[18] Ram Ramrakhya, Dhruv Batra, Erik Wijmans, and Abhishek Das. p. 9

Where the paper says it

org/abs/2109. [18] Ram Ramrakhya, Dhruv Batra, Erik Wijmans, and Abhishek Das.

p. 9
Where the paper says it

org/abs/2109. [18] Ram Ramrakhya, Dhruv Batra, Erik Wijmans, and Abhishek Das.

Pirlnav: Pretraining with imitation and rl finetuning for objectnav. p. 9
Where the paper says it

[18] Ram Ramrakhya, Dhruv Batra, Erik Wijmans, and Abhishek Das. Pirlnav: Pretraining with imitation and rl finetuning for objectnav.

[19] Dhruv Shah, Ajay Sridhar, Nitish Dashora, Kyle Stachowicz, Kevin Black, Noriaki Hirose, and Sergey Levine. p. 9

Where the paper says it

In Computer Vision and Pattern Recognition (CVPR), 2023 IEEE Conference on, 2023. [19] Dhruv Shah, Ajay Sridhar, Nitish Dashora, Kyle Stachowicz, Kevin Black, Noriaki Hirose, and Sergey Levine.

Vint: A foundation model for visual navigation. p. 9
Where the paper says it

[19] Dhruv Shah, Ajay Sridhar, Nitish Dashora, Kyle Stachowicz, Kevin Black, Noriaki Hirose, and Sergey Levine. Vint: A foundation model for visual navigation.

[20] Ajay Sridhar, Dhruv Shah, Catherine Glossop, and Sergey Levine. p. 9

Where the paper says it

press/v229/shah23a. [20] Ajay Sridhar, Dhruv Shah, Catherine Glossop, and Sergey Levine.

NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration. p. 9
Where the paper says it

[20] Ajay Sridhar, Dhruv Shah, Catherine Glossop, and Sergey Levine. NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration.

[21] Boyang Sun, Hanzhi Chen, Stefan Leutenegger, Cesar Cadena, Marc Pollefeys, and Hermann Blum. p. 9

Where the paper says it

org/abs/2310. [21] Boyang Sun, Hanzhi Chen, Stefan Leutenegger, Cesar Cadena, Marc Pollefeys, and Hermann Blum.

Frontiernet: Learning visual cues to explore. p. 9
Where the paper says it

[21] Boyang Sun, Hanzhi Chen, Stefan Leutenegger, Cesar Cadena, Marc Pollefeys, and Hermann Blum. Frontiernet: Learning visual cues to explore.

[22] Qwen Team. Qwen2.5-vl, January 2025. p. 9

Where the paper says it

[22] Qwen Team. 5-vl, January 2025.

[23] Shaoan Wang, Jiazhao Zhang, Minghan Li, Jiahang Liu, Anqi Li, Kui Wu, Fangwei Zhong, Junzhi Yu, Zhizheng Zhang, and He Wang. p. 9

Where the paper says it

io/blog/qwen2. [23] Shaoan Wang, Jiazhao Zhang, Minghan Li, Jiahang Liu, Anqi Li, Kui Wu, Fangwei Zhong, Junzhi Yu, Zhizheng Zhang, and He Wang.

Trackvla: Embodied visual tracking in the wild. p. 9
Where the paper says it

[23] Shaoan Wang, Jiazhao Zhang, Minghan Li, Jiahang Liu, Anqi Li, Kui Wu, Fangwei Zhong, Junzhi Yu, Zhizheng Zhang, and He Wang. Trackvla: Embodied visual tracking in the wild.

[24] Yuncong Yang, Han Yang, Jiachen Zhou, Peihao Chen, Hongxin Zhang, Yilun Du, and Chuang Gan. p. 9

Where the paper says it

org/abs/2505. [24] Yuncong Yang, Han Yang, Jiachen Zhou, Peihao Chen, Hongxin Zhang, Yilun Du, and Chuang Gan.

3d-mem: 3d scene memory for embodied exploration and reasoning. p. 9
Where the paper says it

[24] Yuncong Yang, Han Yang, Jiachen Zhou, Peihao Chen, Hongxin Zhang, Yilun Du, and Chuang Gan. 3d-mem: 3d scene memory for embodied exploration and reasoning.

[25] Zecheng Yin, Chonghao Cheng, and Lizhen. p. 9

Where the paper says it

In Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), pages 17294–17303, June 2025. [25] Zecheng Yin, Chonghao Cheng, and Lizhen.

Navigation with vlm framework: Go to any language. p. 9
Where the paper says it

[25] Zecheng Yin, Chonghao Cheng, and Lizhen. Navigation with vlm framework: Go to any language, 2024.

[26] Zecheng Yin, Hao Zhao, and Zhen Li. p. 9

Where the paper says it

org/abs/2410. [26] Zecheng Yin, Hao Zhao, and Zhen Li.

Hypernav: Hybrid perception for object-oriented navigation in unknown environment. p. 9
Where the paper says it

[26] Zecheng Yin, Hao Zhao, and Zhen Li. Hypernav: Hybrid perception for object-oriented navigation in unknown environment, 2025.

[27] Naoki Yokoyama, Sehoon Ha, Dhruv Batra, Jiuguang Wang, and Bernadette Bucher. p. 9

Where the paper says it

org/abs/2510. [27] Naoki Yokoyama, Sehoon Ha, Dhruv Batra, Jiuguang Wang, and Bernadette Bucher.

Vlfm: Vision-language frontier maps for zero-shot semantic navigation. p. 9
Where the paper says it

[27] Naoki Yokoyama, Sehoon Ha, Dhruv Batra, Jiuguang Wang, and Bernadette Bucher. Vlfm: Vision-language frontier maps for zero-shot semantic navigation.

[28] Naoki Yokoyama, Ram Ramrakhya, Abhishek Das, Dhruv Batra, and Sehoon Ha. p. 9

Where the paper says it

In International Conference on Robotics and Automation (ICRA), 2024. [28] Naoki Yokoyama, Ram Ramrakhya, Abhishek Das, Dhruv Batra, and Sehoon Ha.

Hm3d-ovon: A dataset and benchmark for open-vocabulary object goal navigation. p. 9
Where the paper says it

[28] Naoki Yokoyama, Ram Ramrakhya, Abhishek Das, Dhruv Batra, and Sehoon Ha. Hm3d-ovon: A dataset and benchmark for open-vocabulary object goal navigation.

[29] Xinyao Yu, Sixian Zhang, Xinhang Song, Xiaorong Qin, and Shuqiang Jiang. p. 9

Where the paper says it

In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024. [29] Xinyao Yu, Sixian Zhang, Xinhang Song, Xiaorong Qin, and Shuqiang Jiang.

Trajectory diffusion for objectgoal navigation. p. 9
Where the paper says it

[29] Xinyao Yu, Sixian Zhang, Xinhang Song, Xiaorong Qin, and Shuqiang Jiang. Trajectory diffusion for objectgoal navigation.

[30] Shuaihang Yuan, Hao Huang, Yu Hao, Congcong Wen, Anthony Tzes, and Yi Fang. p. 9

Where the paper says it

ISBN 9798331314385. [30] Shuaihang Yuan, Hao Huang, Yu Hao, Congcong Wen, Anthony Tzes, and Yi Fang.

Gamap: zero-shot object goal navigation with multi-scale geometric-affordance guidance. p. 9
Where the paper says it

[30] Shuaihang Yuan, Hao Huang, Yu Hao, Congcong Wen, Anthony Tzes, and Yi Fang. Gamap: zero-shot object goal navigation with multi-scale geometric-affordance guidance.

[31] Yiming Zeng, Hao Ren, Shuhang Wang, Junlong Huang, and Hui Cheng. p. 9

Where the paper says it

ISBN 9798331314385. [31] Yiming Zeng, Hao Ren, Shuhang Wang, Junlong Huang, and Hui Cheng.

Navidiffusor: Cost-guided diffusion model for visual navigation. p. 9
Where the paper says it

[31] Yiming Zeng, Hao Ren, Shuhang Wang, Junlong Huang, and Hui Cheng. Navidiffusor: Cost-guided diffusion model for visual navigation.

[32] Chaoning Zhang, Dongshen Han, Yu Qiao, Jung Uk Kim, Sung-Ho Bae, Seungkyu Lee, and Choong Seon Hong. p. 9

Where the paper says it

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2025. [32] Chaoning Zhang, Dongshen Han, Yu Qiao, Jung Uk Kim, Sung-Ho Bae, Seungkyu Lee, and Choong Seon Hong.

Faster segment anything: Towards lightweight sam for mobile applications. p. 9
Where the paper says it

[32] Chaoning Zhang, Dongshen Han, Yu Qiao, Jung Uk Kim, Sung-Ho Bae, Seungkyu Lee, and Choong Seon Hong. Faster segment anything: Towards lightweight sam for mobile applications.

[33] Jiazhao Zhang, Anqi Li, Yunpeng Qi, Minghan Li, Jiahang Liu, Shaoan Wang, Haoran Liu, Gengze Zhou, Yuze Wu, Xingxing Li. p. 9

Where the paper says it

14289, 2023. [33] Jiazhao Zhang, Anqi Li, Yunpeng Qi, Minghan Li, Jiahang Liu, Shaoan Wang, Haoran Liu, Gengze Zhou, Yuze Wu, Xingxing Li, et al.

Embodied navigation foundation model. p. 0
Where the paper says it

To equip agents with intelligent exploration capabilities, three main lines of research have emerged: a) End-to-end foundation models. : This type of method usually integrates diverse navigation tasks such as Embodied Question Answering and Human Following, into one universal foundation model[35, 19, 34, 36] with decent performance across multiple tasks.

[34] Jiazhao Zhang, Kunyu Wang, Shaoan Wang, Minghan Li, Haoran Liu, Songlin Wei, Zhongyuan Wang, Zhizheng Zhang, and He Wang. p. 9

Where the paper says it

12129, 2025. [34] Jiazhao Zhang, Kunyu Wang, Shaoan Wang, Minghan Li, Haoran Liu, Songlin Wei, Zhongyuan Wang, Zhizheng Zhang, and He Wang.

Uni-navid: A video-based visionlanguage-action model for unifying embodied navigation tasks. p. 9
Where the paper says it

[34] Jiazhao Zhang, Kunyu Wang, Shaoan Wang, Minghan Li, Haoran Liu, Songlin Wei, Zhongyuan Wang, Zhizheng Zhang, and He Wang. Uni-navid: A video-based visionlanguage-action model for unifying embodied navigation tasks.

[35] Lingfeng Zhang, Xiaoshuai Hao, Yingbo Tang, Haoxiang Fu, Xinyu Zheng, Pengwei Wang, Zhongyuan Wang, Wenbo Ding, and Shanghang Zhang. p. 9

Where the paper says it

Robotics: Science and Systems, 2025. [35] Lingfeng Zhang, Xiaoshuai Hao, Yingbo Tang, Haoxiang Fu, Xinyu Zheng, Pengwei Wang, Zhongyuan Wang, Wenbo Ding, and Shanghang Zhang.

nava3: Understanding any instruction. p. 9
Where the paper says it

[35] Lingfeng Zhang, Xiaoshuai Hao, Yingbo Tang, Haoxiang Fu, Xinyu Zheng, Pengwei Wang, Zhongyuan Wang, Wenbo Ding, and Shanghang Zhang. nava3 : Understanding any instruction, navigating anywhere, finding anything, 2025.

[36] Yufeng Zhong, Chengjian Feng, Feng Yan, Fanfan Liu, Liming Zheng, and Lin Ma. p. 9

Where the paper says it

org/abs/2508. [36] Yufeng Zhong, Chengjian Feng, Feng Yan, Fanfan Liu, Liming Zheng, and Lin Ma.

Robotron-nav: A unified framework for embodied navigation integrating perception, planning and prediction. p. 9
Where the paper says it

[36] Yufeng Zhong, Chengjian Feng, Feng Yan, Fanfan Liu, Liming Zheng, and Lin Ma. Robotron-nav: A unified framework for embodied navigation integrating perception, planning, and prediction.

[4] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng. p. 8

Where the paper says it

In International Conference on Computer Vision (ICCV), 2025. [4] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng, and Yinda Zhang.

Matterport3D: Learning from RGB-D data in indoor environments.

[17] Santhosh K. Ramakrishnan, Aaron Gokaslan, Erik Wijmans, Oleksandr Maksymets, Alex Clegg, John Turner, Eric Undersander, Wojciech Galuba. On evaluation of embodied navigation agents.

[18] Ram Ramrakhya, Dhruv Batra, Erik Wijmans, and Abhishek Das.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Depth-aware intelligent exploration selection (Module a) allows for prioritizing areas based on depth thresholds (e.g., greater than 5m) and using RGB information from candidate regions to provide contextual cues for VLM reasoning, formulated as a Vision Question Answering task like “To find the bed, which of the given areas is the best to reach?”

  2. Global consistency verification (Module b) enhances decision-making by projecting a selected egocentric region onto a top-down map and using it in a second VQA query such as “Is the colored area a good option to find the bed?” This evaluates spatial reasonableness from a global perspective.

  3. History-aware pruning mechanism prevents redundant exploration by removing candidate regions that lie within a predefined distance of the explored area, ensuring regions beyond this margin are considered already sufficiently explored.

  4. Destination Confirmation (Module c) iteratively dilates destination masks if initial candidates are unreachable, allowing the robot to approach as closely as possible to the intended target while maintaining feasibility.

  5. A dynamic path planning module using A-star search computes collision-free trajectories once a confirmed exploration point is reached, and pre-processing obstacle regions by dilating them helps avoid collisions or stepping into areas where structural elements are absent.

  6. Finalization and Termination refines detection by segmenting the object using MobileSAM to create an instance mask, which is combined with depth data and projected onto the top-down map to define the final goal region before navigation commences.

  7. The framework can be extended to memory-augmented navigation on GOAT-BENCH by incorporating a module that records previously found objects, allowing the agent to navigate target object sequences by checking if the next target exists in memory first.

  8. The system achieves a balance between efficiency and generalization, as shown by its performance ranking second on HM3D and first on OVON with a mean EoS of 66.33 across diverse scenes, demonstrating robustness against environmental variations.

Abstract

To locate a target object while exploring the unknown environment is a fundamental capability for autonomous agents, with applications ranging from search-and-rescue to field robots. A simplified version of such task is Object Goal Navigation (ObjNav). In ObjNav, successful arrival at the target object provides a basic measure of performance; however, the efficiency of the navigation trajectory is equally important, as it indicates how intelligently the agent explores and how much time remains for subsequent tasks. In unknown environments, the key to efficient navigation lies in deciding where to explore next. While many prior works aim to address this core challenge and achieved promising performance in certain settings, recent training-based models and non-training frameworks still suffer from generalization and efficiency issues respectively, which in the worst cases can lead to excessive exploration of already-visited areas or redundant back-and-forth motion. We evaluate EffiNav on two widely used simulation benchmarks Habitat Matterport 3D (HM3D) and Open-Vocabulary Object goal Navigation (OVON), and further validate its effectiveness on physical robots in real-world settings. We conduct failure analysis on massive simulation episodes. With minimal modification, we also extend EffiNav to a memory-augmented ObjNav task on the GOAT-BENCH dataset, demonstrating its adaptability beyond standard ObjNav settings. Across two standard metrics--Success Rate (SR) and Success weighted by Path Length (SPL), EffiNav matches or outperforms recent baselines, reflecting its efficiency, robustness, and practical applicability. Recognizing the different emphases of the two datasets, the performances reveals this framework is more balanced and generalizable for efficient ObjNav.

Sources

Related papers