Do Visual-Language Maps Capture Latent Semantics?

Visual-language models (VLMs) have recently been introduced in robotic mapping by using the latent representations, i.e., embeddings, of the VLMs to represent the natural language semantics in the map. The main benefit is moving beyond a small set of human-created labels toward open-vocabulary scene understanding. While there is anecdotal evidence that maps built this way support downstream tasks, such as navigation, rigorous analysis of the quality of the maps using these embeddings is lacking. We investigate two critical properties of map quality: queryability and consistency. The evaluation of queryability addresses the ability to retrieve information from the embeddings. We investigate two aspects of consistency: intra-map consistency and inter-map consistency. Intra-map consistency captures the ability of the embeddings to represent abstract semantic classes, and inter-map consistency captures the generalization properties of the representation. In this paper, we propose a way to analyze the quality of maps created using VLMs, which forms an open-source benchmark to be used when proposing new open-vocabulary map representations. We demonstrate the benchmark by evaluating the maps created by two state-of-the-art methods, VLMaps and OpenScene, using two encoders, LSeg and OpenSeg, using real-world data from the Matterport3D data set. We find that OpenScene outperforms VLMaps with both encoders, and LSeg outperforms OpenSeg with both methods.

Object-oriented mapping in dynamic environments

Grid maps, especially occupancy grid maps, are ubiquitous in many mobile robot applications. To simplify the process of learning the map, grid maps subdivide the world into a grid of cells, whose occupancies are independently estimated using only measurements in the perceptual field of the particular cell. However, the world consists of objects that span multiple cells, which means that measurements falling onto a cell provide evidence on the occupancy of other cells belonging to the same object. This correlation is not captured by current models. In this work, we present a way to generalize the update of grid maps relaxing the assumption of independence by modeling the relationship between the measurements and the occupancy of each cell as a set of latent variables, and jointly estimating those variables and the posterior of the map. Additionally, we propose a method to estimate the latent variables by clustering based on semantic labels and an extension to the Normal Distributions Transfer Occupancy Map (NDT-OM) to facilitate the proposed map update method. We perform comprehensive experiments of map creation and localization with real world data sets, and show that the proposed method creates better maps in highly dynamic environments compared to state-of-the-art methods. Finally, we demonstrate the ability of the proposed method to remove occluded objects from the map in a lifelong map update scenario.

Localization under consistent assumptions over dynamics

Accurate maps are a prerequisite for virtually all autonomous vehicle tasks. Most state-of-the-art maps assume a static world, and therefore dynamic objects are filtered out of the measurements. However, this division ignores movable but non-moving, i.e. semi-static, objects, which are usually recorded in the map and treated as static objects, violating the static world assumption, causing error in the localization. In this paper, we present a method for modeling moving and movable objects for matching the map and the measurements consistently. This reduces the error resulting from inconsistent categorization and treatment of non-static measurements. A semantic segmentation network is used to categorize the measurements into static and semi-static classes, and a background subtraction-based filtering method is used to remove dynamic measurements. Experimental comparison against a state-of-the-art baseline solution using real-world data from Oxford Radar RobotCar data set shows that consistent assumptions over dynamics increase localization accuracy.

Constrained Generative Sampling of 6-DoF Grasps

Most state-of-the-art data-driven grasp sampling methods propose stable and collision-free grasps uniformly on the target object. For bin-picking, executing any of those grasps is sufficient. However, for completing specific tasks, such as squeezing out liquid from a bottle, we want the grasp to be on a specific part on the object body while avoiding other locations, such as the cap. In this work, we present a generative grasp sampling network, VCGS, capable of constrained 6-Degrees-of-Freedom (DoF) grasp sampling. In addition, we also curate a new dataset designed to train and evaluate methods for constrained grasping. The new dataset, called CONG, consists of over 14 million training samples of synthetically rendered point clouds and grasps at random target areas on 2889 objects. VCGS is benchmarked against GraspNet, a state-of-the-art unconstrained grasp sampler, in simulation and on a real robot. The results demonstrate that VCGS achieves a 10-15% higher grasp success rate than the baseline while being 2-3 times as sample efficient.

LSVL: Large-scale season-invariant visual localization for UAVs

Localization of autonomous unmanned aerial vehicles (UAVs) relies heavily on Global Navigation Satellite Systems (GNSS), which are susceptible to interference. Especially in security applications, robust localization algorithms independent of GNSS are needed to provide dependable operations of autonomous UAVs also in interfered conditions. Typical non-GNSS visual localization approaches rely on known starting pose, work only on a small-sized map, or require known flight paths before a mission starts. We consider the problem of localization with no information on initial pose or planned flight path. We propose a solution for global visual localization on a map at scale up to 100 km2, based on matching orthoprojected UAV images to satellite imagery using learned season-invariant descriptors. We show that the method is able to determine heading, latitude and longitude of the UAV at 12.6-18.7 m lateral translation error in as few as 23.2-44.4 updates from an uninformed initialization, also in situations of significant seasonal appearance difference (winter-summer) between the UAV image and the map. We evaluate the characteristics of multiple neural network architectures for generating the descriptors, and likelihood estimation methods that are able to provide fast convergence and low localization error. We also evaluate the operation of the algorithm using real UAV data and evaluate running time on a real-time embedded platform. We believe this is the first work that is able to recover the pose of an UAV at this scale and rate of convergence, while allowing significant seasonal difference between camera observations and map.

Generating people flow from architecture of real unseen environments

Mapping people dynamics is a crucial skill, because it enables robots to coexist in human-inhabited environments. However, learning a model of people dynamics is a time consuming process which requires observation of large amount of people moving in an environment. Moreover, approaches for mapping dynamics are unable to transfer the learned models across environments: each model only able to describe the dynamics of the environment it has been built in. However, the effect of architectural geometry on people movement can be used to estimate their dynamics, and recent work has looked into learning maps of dynamics from geometry. So far however, these methods have evaluated their performance only on small-size synthetic data, leaving the actual ability of these approaches to generalize to real conditions unexplored. In this work we propose a novel approach to learn people dynamics from geometry, where a model is trained and evaluated on real human trajectories in large-scale environments. We then show the ability of our method to generalize to unseen environments, which is unprecedented for maps of dynamics.

Season-invariant GNSS-denied visual localization for UAVs

Knowing the position and orientation of an UAV without GNSS is a critical functionality in autonomous operations of UAVs. Vision-based localization on a known map can be an effective solution, but it is burdened by two main problems: places have different appearance depending on weather and season and the perspective discrepancy between the UAV camera image and the map make matching hard. In this work, we propose a localization solution relying on matching of UAV camera images to georeferenced orthophotos with a trained CNN model that is invariant to significant seasonal appearance difference (winter-summer) between the camera image and map. We compare the convergence speed and localization accuracy of our solution to three other commonly used methods. The results show major improvements with respect to reference methods, especially under high seasonal variation. We finally demonstrate the ability of the method to successfully localize a real UAV, showing that the proposed method is robust to perspective changes.

GNSS-denied geolocalization of UAVs by visual matching of onboard camera images with orthophotos

Localization of low-cost unmanned aerial vehicles(UAVs) often relies on Global Navigation Satellite Systems (GNSS). GNSS are susceptible to both natural disruptions to radio signal and intentional jamming and spoofing by an adversary. A typical way to provide georeferenced localization without GNSS for small UAVs is to have a downward-facing camera and match camera images to a map. The downward-facing camera adds cost, size, and weight to the UAV platform and the orientation limits its usability for other purposes. In this work, we propose a Monte-Carlo localization method for georeferenced localization of an UAV requiring no infrastructure using only inertial measurements, a camera facing an arbitrary direction, and an orthoimage map. We perform orthorectification of the UAV image, relying on a local planarity assumption of the environment, relaxing the requirement of downward-pointing camera. We propose a measure of goodness for the matching score of an orthorectified UAV image and a map. We demonstrate that the system is able to localize globally an UAV with modest requirements for initialization and map resolution.

Augmented Environment Representations with Complete Object Models

While 2D occupancy maps commonly used in mobile robotics enable safe navigation in indoor environments, in order for robots to understand their environment to the level required for them to perform more advanced tasks, representing 3D geometry and semantic environment information is required. We propose a pipeline that can generate a multi-layer representation of indoor environments for robotic applications. The proposed representation includes 3D metric-semantic layers, a 2D occupancy layer, and an object instance layer where known objects are replaced with an approximate model obtained through a novel model-matching approach. The metric-semantic layer and the object instance layer are combined to form an augmented representation of the environment. Experiments show that the proposed shape matching method outperforms a state-of-the-art deep learning method when tasked to complete unseen parts of objects in the scene. The pipeline performance translates well from simulation to real world as shown by F1-score analysis, with semantic segmentation accuracy using Mask R-CNN acting as the major bottleneck. Finally, we also demonstrate on a real robotic platform how the multi-layer map can be used to improve navigation safety.

DDGC: Generative Deep Dexterous Grasping in Clutter

Recent advances in multi-fingered robotic grasping have enabled fast 6-Degrees-Of-Freedom (DOF) single object grasping. Multi-finger grasping in cluttered scenes, on the other hand, remains mostly unexplored due to the added difficulty of reasoning over obstacles which greatly increases the computational time to generate high-quality collision-free grasps. In this work we address such limitations by introducing DDGC, a fast generative multi-finger grasp sampling method that can generate high quality grasps in cluttered scenes from a single RGB-D image. DDGC is built as a network that encodes scene information to produce coarse-to-fine collision-free grasp poses and configurations. We experimentally benchmark DDGC against the simulated-annealing planner in GraspIt! on 1200 simulated cluttered scenes and 7 real world scenes. The results show that DDGC outperforms the baseline on synthesizing high-quality grasps and removing clutter while being 5 times faster. This, in turn, opens up the door for using multi-finger grasps in practical applications which has so far been limited due to the excessive computation time needed by other methods.