Target-driven visual navigation aims at navigating an agent towards a given target based on the observation of the agent. In this task, it is critical to learn informative visual representation and robust navigation policy. Aiming to improve these two components, this paper proposes three complementary techniques, object relation graph (ORG), trial-driven imitation learning (IL), and a memory-augmented tentative policy network (TPN). ORG improves visual representation learning by integrating object relationships, including category closeness and spatial correlations, e.g., a TV usually co-occurs with a remote spatially. Both Trial-driven IL and TPN underlie robust navigation policy, instructing the agent to escape from deadlock states, such as looping or being stuck. Specifically, trial-driven IL is a type of supervision used in policy network training, while TPN, mimicking the IL supervision in unseen environment, is applied in testing. Experiment in the artificial environment AI2-Thor validates that each of the techniques is effective. When combined, the techniques bring significantly improvement over baseline methods in navigation effectiveness and efficiency in unseen environments. We report 22.8% and 23.5% increase in success rate and Success weighted by Path Length (SPL), respectively. The code is available at https://github.com/xiaobaishu0097/ECCV-VN.git.
Monocular visual navigation methods have seen significant advances in the last decade, recently producing several real-time solutions for autonomously navigating small unmanned aircraft systems without relying on GPS. This is critical for military operations which may involve environments where GPS signals are degraded or denied. However, testing and comparing visual navigation algorithms remains a challenge since visual data is expensive to gather. Conducting flight tests in a virtual environment is an attractive solution prior to committing to outdoor testing. This work presents a virtual testbed for conducting simulated flight tests over real-world terrain and analyzing the real-time performance of visual navigation algorithms at 31 Hz. This tool was created to ultimately find a visual odometry algorithm appropriate for further GPS-denied navigation research on fixed-wing aircraft, even though all of the algorithms were designed for other modalities. This testbed was used to evaluate three current state-of-the-art, open-source monocular visual odometry algorithms on a fixed-wing platform: Direct Sparse Odometry, Semi-Direct Visual Odometry, and ORB-SLAM2 (with loop closures disabled).
Semantic cues and statistical regularities in real-world environment layouts can improve efficiency for navigation in novel environments. This paper learns and leverages such semantic cues for navigating to objects of interest in novel environments, by simply watching YouTube videos. This is challenging because YouTube videos don't come with labels for actions or goals, and may not even showcase optimal behavior. Our method tackles these challenges through the use of Q-learning on pseudo-labeled transition quadruples (image, action, next image, reward). We show that such off-policy Q-learning from passive data is able to learn meaningful semantic cues for navigation. These cues, when used in a hierarchical navigation policy, lead to improved efficiency at the ObjectGoal task in visually realistic simulations. We observe a relative improvement of 15-83% over end-to-end RL, behavior cloning, and classical methods, while using minimal direct interaction.
In this paper, we propose a dense monocular SLAM system, named DeepRelativeFusion, that is capable to recover a globally consistent 3D structure. To this end, we use a visual SLAM algorithm to reliably recover the camera poses and semi-dense depth maps of the keyframes, and then use relative depth prediction to densify the semi-dense depth maps and refine the keyframe pose-graph. To improve the semi-dense depth maps, we propose an adaptive filtering scheme, which is a structure-preserving weighted average smoothing filter that takes into account the pixel intensity and depth of the neighbouring pixels, yielding substantial reconstruction accuracy gain in densification. To perform densification, we introduce two incremental improvements upon the energy minimization framework proposed by DeepFusion: (1) an improved cost function, and (2) the use of single-image relative depth prediction. After densification, we update the keyframes with two-view consistent optimized semi-dense and dense depth maps to improve pose-graph optimization, providing a feedback loop to refine the keyframe poses for accurate scene reconstruction. Our system outperforms the state-of-the-art dense SLAM systems quantitatively in dense reconstruction accuracy by a large margin.
This paper studies the problem of image-goal navigation which involves navigating to the location indicated by a goal image in a novel previously unseen environment. To tackle this problem, we design topological representations for space that effectively leverage semantics and afford approximate geometric reasoning. At the heart of our representations are nodes with associated semantic features, that are interconnected using coarse geometric information. We describe supervised learning-based algorithms that can build, maintain and use such representations under noisy actuation. Experimental study in visually and physically realistic simulation suggests that our method builds effective representations that capture structural regularities and efficiently solve long-horizon navigation problems. We observe a relative improvement of more than 50% over existing methods that study this task.
Training deep reinforcement learning agents on environments with multiple levels / scenes / conditions from the same task, has become essential for many applications aiming to achieve generalization and domain transfer from simulation to the real world. While such a strategy is helpful with generalization, the use of multiple scenes significantly increases the variance of samples collected for policy gradient computations. Current methods continue to view this collection of scenes as a single Markov Decision Process (MDP) with a common value function; however, we argue that it is better to treat the collection as a single environment with multiple underlying MDPs. To this end, we propose a dynamic value estimation (DVE) technique for these multiple-MDP environments, motivated by the clustering effect observed in the value function distribution across different scenes. The resulting agent is able to learn a more accurate and scene-specific value function estimate (and hence the advantage function), leading to a lower sample variance. Our proposed approach is simple to accommodate with several existing implementations (like PPO, A3C) and results in consistent improvements for a range of ProcGen environments and the AI2-THOR framework based visual navigation task.
Embodied artificial intelligence (AI) tasks shift from tasks focusing on internet images to active settings involving embodied agents that perceive and act within 3D environments. In this paper, we investigate the target-driven visual navigation using deep reinforcement learning (DRL) in 3D indoor scenes, whose navigation task aims to train an agent that can intelligently make a series of decisions to arrive at a pre-specified target location from any possible starting positions only based on egocentric views. However, most navigation methods currently struggle against several challenging problems, such as data efficiency, automatic obstacle avoidance, and generalization. Generalization problem means that agent does not have the ability to transfer navigation skills learned from previous experience to unseen targets and scenes. To address these issues, we incorporate two designs into classic DRL framework: attention on 3D knowledge graph (KG) and target skill extension (TSE) module. On the one hand, our proposed method combines visual features and 3D spatial representations to learn navigation policy. On the other hand, TSE module is used to generate sub-targets which allow agent to learn from failures. Specifically, our 3D spatial relationships are encoded through recently popular graph convolutional network (GCN). Considering the real world settings, our work also considers open action and adds actionable targets into conventional navigation situations. Those more difficult settings are applied to test whether DRL agent really understand its task, navigating environment, and can carry out reasoning. Our experiments, performed in the AI2-THOR, show that our model outperforms the baselines in both SR and SPL metrics, and improves generalization ability across targets and scenes.
Several animal species (e.g., bats, dolphins, and whales) and even visually impaired humans have the remarkable ability to perform echolocation: a biological sonar used to perceive spatial layout and locate objects in the world. We explore the spatial cues contained in echoes and how they can benefit vision tasks that require spatial reasoning. First we capture echo responses in photo-realistic 3D indoor scene environments. Then we propose a novel interaction-based representation learning framework that learns useful visual features via echolocation. We show that the learned image features are useful for multiple downstream vision tasks requiring spatial reasoning---monocular depth estimation, surface normal estimation, and visual navigation---with results comparable or even better than heavily supervised pre-training. Our work opens a new path for representation learning for embodied agents, where supervision comes from interacting with the physical world.
In this work, we present a method for obtaining an implicit objective function for vision-based navigation. The proposed methodology relies on Imitation Learning, Model Predictive Control (MPC), and an interpretation technique used in Deep Neural Networks. We use Imitation Learning as a means to do Inverse Reinforcement Learning in order to create an approximate cost function generator for a visual navigation challenge. The resulting cost function, the costmap, is used in conjunction with MPC for real-time control and outperforms other state-of-the-art costmap generators in novel environments. The proposed process allows for simple training and robustness to out-of-sample data. We apply our method to the task of vision-based autonomous driving in multiple real and simulated environments and show its generalizability.
We humans can impeccably search for a target object, given its name only, even in an unseen environment. We argue that this ability is largely due to three main reasons: the incorporation of prior knowledge (or experience), the adaptation of it to the new environment using the observed visual cues and most importantly optimistically searching without giving up early. This is currently missing in the state-of-the-art visual navigation methods based on Reinforcement Learning (RL). In this paper, we propose to use externally learned prior knowledge of the relative object locations and integrate it into our model by constructing a neural graph. In order to efficiently incorporate the graph without increasing the state-space complexity, we propose our Graph-based Value Estimation (GVE) module. GVE provides a more accurate baseline for estimating the Advantage function in actor-critic RL algorithm. This results in reduced value estimation error and, consequently, convergence to a more optimal policy. Through empirical studies, we show that our agent, dubbed as the optimistic agent, has a more realistic estimate of the state value during a navigation episode which leads to a higher success rate. Our extensive ablation studies show the efficacy of our simple method which achieves the state-of-the-art results measured by the conventional visual navigation metrics, e.g. Success Rate (SR) and Success weighted by Path Length (SPL), in AI2THOR environment.
We consider the task of underwater robot navigation for the purpose of collecting scientifically relevant video data for environmental monitoring. The majority of field robots that currently perform monitoring tasks in unstructured natural environments navigate via path-tracking a pre-specified sequence of waypoints. Although this navigation method is often necessary, it is limiting because the robot does not have a model of what the scientist deems to be relevant visual observations. Thus, the robot can neither visually search for particular types of objects, nor focus its attention on parts of the scene that might be more relevant than the pre-specified waypoints and viewpoints. In this paper we propose a method that enables informed visual navigation via a learned visual similarity operator that guides the robot's visual search towards parts of the scene that look like an exemplar image, which is given by the user as a high-level specification for data collection. We propose and evaluate a weakly supervised video representation learning method that outperforms ImageNet embeddings for similarity tasks in the underwater domain. We also demonstrate the deployment of this similarity operator during informed visual navigation in collaborative environmental monitoring scenarios, in large-scale field trials, where the robot and a human scientist collaboratively search for relevant visual content.
Real world visual navigation requires robots to operate in unfamiliar, human-occupied dynamic environments. Navigation around humans is especially difficult because it requires anticipating their future motion, which can be quite challenging. We propose an approach that combines learning-based perception with model-based optimal control to navigate among humans based only on monocular, first-person RGB images. Our approach is enabled by our novel data-generation tool, HumANav that allows for photorealistic renderings of indoor environment scenes with humans in them, which are then used to train the perception module entirely in simulation. Through simulations and experiments on a mobile robot, we demonstrate that the learned navigation policies can anticipate and react to humans without explicitly predicting future human motion, generalize to previously unseen environments and human behaviors, and transfer directly from simulation to reality. Videos describing our approach and experiments, as well as a demo of HumANav are available on the project website.
Direct search for objects as part of navigation poses a challenge for small items. Utilizing context in the form of object-object relationships enable hierarchical search for targets efficiently. Most of the current approaches tend to directly incorporate sensory input into a reward-based learning approach, without learning about object relationships in the natural environment, and thus generalize poorly across domains. We present Memory-utilized Joint hierarchical Object Learning for Navigation in Indoor Rooms (MJOLNIR), a target-driven navigation algorithm, which considers the inherent relationship between target objects, and the more salient contextual objects occurring in its surrounding. Extensive experiments conducted across multiple environment settings show an $82.9\%$ and $93.5\%$ gain over existing state-of-the-art navigation methods in terms of the success rate (SR), and success weighted by path length (SPL), respectively. We also show that our model learns to converge much faster than other algorithms, without suffering from the well-known overfitting problem. Additional details regarding the supplementary material and code are available at https://sites.google.com/eng.ucsd.edu/mjolnir.
To operate effectively in the real world, agents should be able to act from high-dimensional raw sensory input such as images and achieve diverse goals across long time-horizons. Current deep reinforcement and imitation learning methods can learn directly from high-dimensional inputs but do not scale well to long-horizon tasks. In contrast, classical graphical methods like A* search are able to solve long-horizon tasks, but assume that the state space is abstracted away from raw sensory input. Recent works have attempted to combine the strengths of deep learning and classical planning; however, dominant methods in this domain are still quite brittle and scale poorly with the size of the environment. We introduce Sparse Graphical Memory (SGM), a new data structure that stores states and feasible transitions in a sparse memory. SGM aggregates states according to a novel two-way consistency objective, adapting classic state aggregation criteria to goal-conditioned RL: two states are redundant when they are interchangeable both as goals and as starting states. Theoretically, we prove that merging nodes according to two-way consistency leads to an increase in shortest path lengths that scales only linearly with the merging threshold. Experimentally, we show that SGM significantly outperforms current state of the art methods on long horizon, sparse-reward visual navigation tasks. Project video and code are available at https://mishalaskin.github.io/sgm/
This paper addresses the problem of building augmented metric representations of scenes with semantic information from RGB-D images. We propose a complete framework to create an enhanced map representation of the environment with object-level information to be used in several applications such as human-robot interaction, assistive robotics, visual navigation, or in manipulation tasks. Our formulation leverages a CNN-based object detector (Yolo) with a 3D model-based segmentation technique to perform instance semantic segmentation, and to localize, identify, and track different classes of objects in the scene. The tracking and positioning of semantic classes is done with a dictionary of Kalman filters in order to combine sensor measurements over time and then providing more accurate maps. The formulation is designed to identify and to disregard dynamic objects in order to obtain a medium-term invariant map representation. The proposed method was evaluated with collected and publicly available RGB-D data sequences acquired in different indoor scenes. Experimental results show the potential of the technique to produce augmented semantic maps containing several objects (notably doors). We also provide to the community a dataset composed of annotated object classes (doors, fire extinguishers, benches, water fountains) and their positioning, as well as the source code as ROS packages.
Autonomous navigation emerges from both motion and local visual perception in real-world environments. However, most successful robotic motion estimation methods (e.g. VO, SLAM, SfM) and vision systems (e.g. CNN, visual place recognition-VPR) are often separately used for mapping and localization tasks. Conversely, recent reinforcement learning (RL) based methods for visual navigation rely on the quality of GPS data reception, which may not be reliable when directly using it as ground truth across multiple, month-spaced traversals in large environments. In this paper, we propose a novel motion and visual perception approach, dubbed MVP, that unifies these two sensor modalities for large-scale, target-driven navigation tasks. Our MVP-based method can learn faster, and is more accurate and robust to both extreme environmental changes and poor GPS data than corresponding vision-only navigation methods. MVP temporally incorporates compact image representations, obtained using VPR, with optimized motion estimation data, including but not limited to those from VO or optimized radar odometry (RO), to efficiently learn self-supervised navigation policies via RL. We evaluate our method on two large real-world datasets, Oxford Robotcar and Nordland Railway, over a range of weather (e.g. overcast, night, snow, sun, rain, clouds) and seasonal (e.g. winter, spring, fall, summer) conditions using the new CityLearn framework; an interactive environment for efficiently training navigation agents. Our experimental results, on traversals of the Oxford RobotCar dataset with no GPS data, show that MVP can achieve 53% and 93% navigation success rate using VO and RO, respectively, compared to 7% for a vision-only method. We additionally report a trade-off between the RL success rate and the motion estimation precision.
Visual Indoor Navigation (VIN) task has drawn increasing attention from the data-driven machine learning communities especially with the recently reported success from learning-based methods. Due to the innate complexity of this task, researchers have tried approaching the problem from a variety of different angles, the full scope of which has not yet been captured within an overarching report. This survey first summarizes the representative work of learning-based approaches for the VIN task and then identifies and discusses lingering issues impeding the VIN performance, as well as motivates future research in these key areas worth exploring for the community.
Learning to navigate in a visual environment following natural-language instructions is a challenging task, because the multimodal inputs to the agent are highly variable, and the training data on a new task is often limited. In this paper, we present the first pre-training and fine-tuning paradigm for vision-and-language navigation (VLN) tasks. By training on a large amount of image-text-action triplets in a self-supervised learning manner, the pre-trained model provides generic representations of visual environments and language instructions. It can be easily used as a drop-in for existing VLN frameworks, leading to the proposed agent called Prevalent. It learns more effectively in new tasks and generalizes better in a previously unseen environment. The performance is validated on three VLN tasks. On the Room-to-Room benchmark, our model improves the state-of-the-art from 47% to 51% on success rate weighted by path length. Further, the learned representation is transferable to other VLN tasks. On two recent tasks, vision-and-dialog navigation and "Help, Anna!" the proposed Prevalent leads to significant improvement over existing methods, achieving a new state of the art.
Deep reinforcement learning is successful in decision making for sophisticated games, such as Atari, Go, etc. However, real-world decision making often requires reasoning with partial information extracted from complex visual observations. This paper presents Discriminative Particle Filter Reinforcement Learning (DPFRL), a new reinforcement learning framework for complex partial observations. DPFRL encodes a differentiable particle filter in the neural network policy for explicit reasoning with partial observations over time. The particle filter maintains a belief using learned discriminative update, which is trained end-to-end for decision making. We show that using the discriminative update instead of standard generative models results in significantly improved performance, especially for tasks with complex visual observations, because they circumvent the difficulty of modeling complex observations that are irrelevant to decision making. In addition, to extract features from the particle belief, we propose a new type of belief feature based on the moment generating function. DPFRL outperforms state-of-the-art POMDP RL models in Flickering Atari Games, an existing POMDP RL benchmark, and in Natural Flickering Atari Games, a new, more challenging POMDP RL benchmark introduced in this paper. Further, DPFRL performs well for visual navigation with real-world data in the Habitat environment.
Autonomous systems possess the features of inferring their own state, understanding their surroundings, and performing autonomous navigation. With the applications of learning systems, like deep learning and reinforcement learning, the visual-based self-state estimation, environment perception and navigation capabilities of autonomous systems have been efficiently addressed, and many new learning-based algorithms have surfaced with respect to autonomous visual perception and navigation. In this review, we focus on the applications of learning-based monocular approaches in ego-motion perception, environment perception and navigation in autonomous systems, which is different from previous reviews that discussed traditional methods. First, we delineate the shortcomings of existing classical visual simultaneous localization and mapping (vSLAM) solutions, which demonstrate the necessity to integrate deep learning techniques. Second, we review the visual-based environmental perception and understanding methods based on deep learning, including deep learning-based monocular depth estimation, monocular ego-motion prediction, image enhancement, object detection, semantic segmentation, and their combinations with traditional vSLAM frameworks. Then, we focus on the visual navigation based on learning systems, mainly including reinforcement learning and deep reinforcement learning. Finally, we examine several challenges and promising directions discussed and concluded in related research of learning systems in the era of computer science and robotics.