Frank Dellaert

We propose a new metric for robot state estimation based on the recently introduced Lie group definition. Our metric is related to prior metrics for SLAM but explicitly takes into account the linear velocity of the state estimate, improving over current pose-based trajectory analysis. This has the benefit of providing a single, quantitative metric to evaluate state estimation algorithms against, while being compatible with existing tools and libraries. Since ground truth data generally consists of pose data from motion capture systems, we also propose an approach to compute the ground truth linear velocity based on polynomial interpolation. Using Chebyshev interpolation and a pseudospectral parameterization, we can accurately estimate the ground truth linear velocity of the trajectory in an optimal fashion with best approximation error. We demonstrate how this approach performs on multiple robotic platforms where accurate state estimation is vital, and compare it to alternative approaches such as finite differences. The pseudospectral parameterization also provides a means of trajectory data compression as an additional benefit. Experimental results show our method provides a valid and accurate means of comparing state estimation systems, which is also easy to interpret and report.

Robot art presents an opportunity to both showcase and advance state-of-the-art robotics through the challenging task of creating art. Creating large-scale artworks in particular engages the public in a way that small-scale works cannot, and the distinct qualities of brush strokes contribute to an organic and human-like quality. Combining the large scale of murals with the strokes of the brush medium presents an especially impactful result, but also introduces unique challenges in maintaining precise, dextrous motion control of the brush across such a large workspace. In this work, we present the first robot to our knowledge that can paint architectural-scale murals with a brush. We create a hybrid robot consisting of a cable-driven parallel robot and 4 degree of freedom (DoF) serial manipulator to paint a 27m by 3.7m mural on windows spanning 2-stories of a building. We discuss our approach to achieving both the scale and accuracy required for brush-painting a mural through a combination of novel mechanical design elements, coordinated planning and control, and on-site calibration algorithms with experimental validations.

While initial approaches to Structure-from-Motion (SfM) revolved around both global and incremental methods, most recent applications rely on incremental systems to estimate camera poses due to their superior robustness. Though there has been tremendous progress in SfM `front-ends' powered by deep models learned from data, the state-of-the-art (incremental) SfM pipelines still rely on classical SIFT features, developed in 2004. In this work, we investigate whether leveraging the developments in feature extraction and matching helps global SfM perform on par with the SOTA incremental SfM approach (COLMAP). To do so, we design a modular SfM framework that allows us to easily combine developments in different stages of the SfM pipeline. Our experiments show that while developments in deep-learning based two-view correspondence estimation do translate to improvements in point density for scenes reconstructed with global SfM, none of them outperform SIFT when comparing with incremental SfM results on a range of datasets. Our SfM system is designed from the ground up to leverage distributed computation, enabling us to parallelize computation on multiple machines and scale to large scenes.

Trajectory retiming is the task of computing a feasible time parameterization to traverse a path. It is commonly used in the decoupled approach to trajectory optimization whereby a path is first found, then a retiming algorithm computes a speed profile that satisfies kino-dynamic and other constraints. While trajectory retiming is most often formulated with the minimum-time objective (i.e. traverse the path as fast as possible), it is not always the most desirable objective, particularly when we seek to balance multiple objectives or when bang-bang control is unsuitable. In this paper, we present a novel algorithm based on factor graph variable elimination that can solve for the global optimum of the retiming problem with quadratic objectives as well (e.g. minimize control effort or match a nominal speed by minimizing squared error), which may extend to arbitrary objectives with iteration. Our work extends prior works, which find only solutions on the boundary of the feasible region, while maintaining the same linear time complexity from a single forward-backward pass. We experimentally demonstrate that (1) we achieve better real-world robot performance by using quadratic objectives in place of the minimum-time objective, and (2) our implementation is comparable or faster than state-of-the-art retiming algorithms.

We propose a manifold optimization approach for solving constrained inference and planning problems. The approach employs a framework that transforms an arbitrary nonlinear equality constrained optimization problem into an unconstrained manifold optimization problem. The core of the transformation process is the formulation of constraint manifolds that represent sets of variables subject to equality constraints. We propose various approaches to define the tan-gent spaces and retraction operations of constraint manifolds, which are crucial for manifold optimization. We evaluate our constraint manifold optimization approach on multiple constrained inference and planning problems, and show that it generates strictly feasible results with increased efficiency as compared to state-of-the-art constrained optimization methods.

We propose a novel hybrid cable-based robot with manipulator and camera for high-accuracy, medium-throughput plant monitoring in a vertical hydroponic farm and, as an example application, demonstrate non-destructive plant mass estimation. Plant monitoring with high temporal and spatial resolution is important to both farmers and researchers to detect anomalies and develop predictive models for plant growth. The availability of high-quality, off-the-shelf structure-from-motion (SfM) and photogrammetry packages has enabled a vibrant community of roboticists to apply computer vision for non-destructive plant monitoring. While existing approaches tend to focus on either high-throughput (e.g. satellite, unmanned aerial vehicle (UAV), vehicle-mounted, conveyor-belt imagery) or high-accuracy/robustness to occlusions (e.g. turn-table scanner or robot arm), we propose a middle-ground that achieves high accuracy …

For navigation problems involving dead reckoning of odometry measurements aided with additional sensors, we introduce a method that treats the lateral components of the odometry measurements as constraints, thereby reducing the dimensionality of the state representation. The constrained lateral motion approach is best suited for factor graph representations of ground vehicle and fixed-wing aerial vehicle navigation, whereby the tangential component of motion is typically much greater than the lateral component. We conduct experiments in both 2D and 3D cooperative navigation scenarios aided by inter-vehicle range measurements, and show that we achieve faster convergence with more efficient optimization with our new parameterization.

We propose factor graph optimization for simultaneous planning, control, and trajectory estimation for collision-free navigation of autonomous systems in environments with moving objects. The proposed online probabilistic motion planning and trajectory estimation navigation technique generates optimal collision-free state and control trajectories for autonomous vehicles when the obstacle motion model is both unknown and known. We evaluate the utility of the algorithm to support future autonomous robotic space missions.

We propose a new system for automatic 2D floorplan reconstruction that is enabled by SALVe, our novel pairwise learned alignment verifier. The inputs to our system are sparsely located 360 panoramas, whose semantic features (windows, doors, and openings) are inferred and used to hypothesize pairwise room adjacency or overlap. SALVe initializes a pose graph, which is subsequently optimized using GTSAM . Once the room poses are computed, room layouts are inferred using HorizonNet , and the floorplan is constructed by stitching the most confident layout boundaries. We validate our system qualitatively and quantitatively as well as through ablation studies, showing that it outperforms state-of-the-art SfM systems in completeness by over 200%, without sacrificing accuracy. Our results point to the significance of our work: poses of 81% of panoramas are localized in the first 2 connected components (CCs …

Legged robot locomotion is a challenging task due to a myriad of sub-problems, such as the hybrid dynamics of foot contact and the effects of the desired gait on the terrain. Accurate and efficient state estimation of the floating base and the feet joints can help alleviate much of these issues by providing feedback information to robot controllers. Current state estimation methods are highly reliant on a conjunction of visual and inertial measurements to provide real-time estimates, thus being handicapped in perceptually poor environments. In this work, we show that by leveraging the kinematic chain model of the robot via a factor graph formulation, we can perform state estimation of the base and the leg joints using primarily proprioceptive inertial data. We perform state estimation using a combination of preintegrated IMU measurements, forward kinematic computations, and contact detections in a factor-graph based …

We present GTGraffiti, a graffiti painting system from Georgia Tech that tackles challenges in art, hardware, and human-robot collaboration. The problem of painting graffiti in a human style is particularly challenging and requires a system-level approach because the robotics and art must be designed around each other. The robot must be highly dynamic over a large workspace while the artist must work within the robot's limitations. Our approach consists of three stages: artwork capture, robot hardware, and planning & control. We use motion capture to capture collaborator painting motions which are then composed and processed into a time-varying linear feedback controller for a cable-driven parallel robot (CDPR) to execute. In this work, we will describe the capturing process, the design and construction of a purpose-built CDPR, and the software for turning an artist's vision into control commands. Our work …

In this work, we investigate the problem of incre-mentally solving constrained non-linear optimization problems formulated as factor graphs. Prior incremental solvers were either restricted to the unconstrained case or required periodic batch relinearizations of the objective and constraints which are expensive and detract from the online nature of the algorithm. We present InCOpt, an Augmented Lagrangian-based incremental constrained optimizer that views matrix operations as message passing over the Bayes tree. We first show how the linear system, resulting from linearizing the constrained objective, can be represented as a Bayes tree. We then propose an algorithm that views forward and back substitutions, which naturally arise from solving the Lagrangian, as upward and downward passes on the tree. Using this formulation, In-COpt can exploit properties such as fluid/online relinearization leading to …

Quadrupeds are robots that have been of interest in the past few years due to their versatility in navigating across various terrain and utility in several applications. For quadrupeds to navigate without a predefined map a priori, they must rely on SLAM approaches to localize and build the map of the environment. Despite the surge of interest and research development in SLAM and quadrupeds, there still has yet to be an open-source package that capitalizes on the onboard sensors of an affordable quadruped. This motivates the A1 SLAM package, which is an open-source ROS package that provides the Unitree A1 quadruped with real-time, high performing SLAM capabilities using the default sensors shipped with the robot. A1 SLAM solves the PoseSLAM problem using the factor graph paradigm to optimize for the poses throughout the trajectory. A major design feature of the algorithm is using a sliding window of fully connected LiDAR odometry factors. A1 SLAM has been benchmarked against Google's Cartographer and has showed superior performance especially with trajectories experiencing aggressive motion.

This paper introduces a novel proprioceptive state estimator for legged robots based on a learned displacement measurement from IMU data. Recent research in pedestrian tracking has shown that motion can be inferred from inertial data using convolutional neural networks. A learned inertial displacement measurement can improve state estimation in challenging scenarios where leg odometry is unreliable, such as slipping and compressible terrains. Our work learns to estimate a displacement measurement from IMU data which is then fused with traditional leg odometry. Our approach greatly reduces the drift of proprioceptive state estimation, which is critical for legged robots deployed in vision and lidar denied environments such as foggy sewers or dusty mines. We compared results from an EKF and an incremental fixed-lag factor graph estimator using data from several real robot experiments crossing challenging terrains. Our results show a reduction of relative pose error by 37% in challenging scenarios when compared to a traditional kinematic-inertial estimator without learned measurement. We also demonstrate a 22% reduction in error when used with vision systems in visually degraded environments such as an underground mine.

We present a locally optimal tracking controller for Cable Driven Parallel Robot (CDPR) control based on a time-varying Linear Quadratic Gaussian (TV-LQG) controller. In contrast to many methods which use fixed feedback gains, our time-varying controller computes the optimal gains depending on the location in the workspace and the future trajectory. Meanwhile, we rely heavily on offline computation to reduce the burden of online implementation and feasibility checking. Following the growing popularity of probabilistic graphical models for optimal control, we use factor graphs as a tool to formulate our controller for their efficiency, intuitiveness, and modularity. The topology of a factor graph encodes the relevant structural properties of equations in a way that facilitates insight and efficient computation using sparse linear algebra solvers. We first use factor graph optimization to compute a nominal trajectory, then …

Visual Inertial Odometry (VIO) is one of the most established state estimation methods for mobile platforms. However, when visual tracking fails, VIO algorithms quickly diverge due to rapid error accumulation during inertial data integration. This error is typically modeled as a combination of additive Gaussian noise and a slowly changing bias which evolves as a random walk. In this work, we propose to train a neural network to learn the true bias evolution. We implement and compare two common sequential deep learning architectures: LSTMs and Transformers. Our approach follows from recent learning-based inertial estimators, but, instead of learning a motion model, we target IMU bias explicitly, which allows us to generalize to locomotion patterns unseen in training. We show that our proposed method improves state estimation in visually challenging situations across a wide range of motions by quadrupedal …

We present PanopticNeRF, an object-aware neural scene representation that decomposes a scene into a set of objects (things) and background (stuff). Each object is represented by a separate MLP that takes a position, direction, and time and outputs density and radiance. The background is represented by a similar MLP that also outputs semantics. Importantly, the object MLPs are specific to each instance and initialized with meta-learning, and thus can be smaller and faster than previous object-aware approaches, while still leveraging category-specific priors. We propose a system to infer the PanopticNeRF representation from a set of color images. We use off-the-shelf algorithms to predict camera poses, object bounding boxes, object categories, and 2D image semantic segmentations. Then we jointly optimize the MLP weights and bounding box parameters using analysis-by-synthesis with self-supervision from the color images and pseudo-supervision from predicted semantic segmentations. PanopticNeRF can be effectively used for multiple 2D and 3D tasks like 3D scene editing, 3D panoptic reconstruction, novel view and semantic synthesis, 2D panoptic segmentation, and multiview depth prediction. We demonstrate these applications on several difficult, dynamic scenes with moving objects.

We propose im2nerf, a learning framework that predicts a continuous neural object representation given a single input image in the wild, supervised by only segmentation output from off-the-shelf recognition methods. The standard approach to constructing neural radiance fields takes advantage of multi-view consistency and requires many calibrated views of a scene, a requirement that cannot be satisfied when learning on large-scale image data in the wild. We take a step towards addressing this shortcoming by introducing a model that encodes the input image into a disentangled object representation that contains a code for object shape, a code for object appearance, and an estimated camera pose from which the object image is captured. Our model conditions a NeRF on the predicted object representation and uses volume rendering to generate images from novel views. We train the model end-to-end on a large collection of input images. As the model is only provided with single-view images, the problem is highly under-constrained. Therefore, in addition to using a reconstruction loss on the synthesized input view, we use an auxiliary adversarial loss on the novel rendered views. Furthermore, we leverage object symmetry and cycle camera pose consistency. We conduct extensive quantitative and qualitative experiments on the ShapeNet dataset as well as qualitative experiments on Open Images dataset. We show that in all cases, im2nerf achieves the state-of-the-art performance for novel view synthesis from a single-view unposed image in the wild.

We present a manifold optimization approach to solve inference and planning problems with range constraints. The core of our approach is the definition of a manifold that represents points or poses with range constraints. We discover that the manifold of range-constrained points is homogeneous under the rigid transformation group action, and utilize the group action to derive the tangent space, retraction and topology of the manifold. We evaluate the performance of manifold optimization approach on solving range-constrained inference problems over state-of-the-art constrained optimization methods. The results show that manifold optimization with the range-constraint manifold achieves both faster speed and better constraint satisfaction. We further study the conditions of inference problems that we can treat range measurements as constraints in practice.

For the Hilti Challenge 2022, we created two systems, one building upon the other. The first system is FL2BIPS which utilizes the iEKF algorithm FAST-LIO2 and Bayesian ICP PoseSLAM, whereas the second system is GTSFM, a structure from motion pipeline with factor graph backend optimization powered by GTSAM