The small size, high dexterity, and intrinsic compliance of continuum robots (CRs) make them well suited for constrained environments. Solving the inverse kinematics (IK), that is finding robot joint configurations that satisfy desired position or pose queries, is a fundamental challenge in motion planning, control, and calibration for any robot structure. For CRs, the need to avoid obstacles in tightly confined workspaces greatly complicates the search for feasible IK solutions. Without an accurate initialization or multiple re-starts, existing algorithms often fail to find a solution. We present CIDGIKc (Convex Iteration for Distance-Geometric Inverse Kinematics for Continuum Robots), an algorithm that solves these nonconvex feasibility problems with a sequence of semidefinite programs whose objectives are designed to encourage low-rank minimizers. CIDGIKc is enabled by a novel distance-geometric parameterization of constant curvature segment geometry for CRs with extensible segments. The resulting IK formulation involves only quadratic expressions and can efficiently incorporate a large number of collision avoidance constraints. Our experimental results demonstrate >98% solve success rates within complex, highly cluttered environments which existing algorithms cannot account for.
Following four successful years in the SAE AutoDrive Challenge Series I, the University of Toronto is participating in the Series II competition to develop a Level 4 autonomous passenger vehicle capable of handling various urban driving scenarios by 2025. Accurate detection of traffic lights and correct identification of their states is essential for safe autonomous operation in cities. Herein, we describe our recently-redesigned traffic light perception system for autonomous vehicles like the University of Toronto's self-driving car, Artemis. Similar to most traffic light perception systems, we rely primarily on camera-based object detectors. We deploy the YOLOv5 detector for bounding box regression and traffic light classification across multiple cameras and fuse the observations. To improve robustness, we incorporate priors from high-definition semantic maps and perform state filtering using hidden Markov models. We demonstrate a multi-camera, real time-capable traffic light perception pipeline that handles complex situations including multiple visible intersections, traffic light variations, temporary occlusion, and flashing light states. To validate our system, we collected and annotated a varied dataset incorporating flashing states and a range of occlusion types. Our results show superior performance in challenging real-world scenarios compared to single-frame, single-camera object detection.
Advances in lidar technology have made the collection of 3D point clouds fast and easy. While most lidar sensors return per-point intensity (or reflectance) values along with range measurements, flash lidar sensors are able to provide information about the shape of the return pulse. The shape of the return waveform is affected by many factors, including the distance that the light pulse travels and the angle of incidence with a surface. Importantly, the shape of the return waveform also depends on the material properties of the reflecting surface. In this paper, we investigate whether the material type or class can be determined from the full-waveform response. First, as a proof of concept, we demonstrate that the extra information about material class, if known accurately, can improve performance on scene understanding tasks such as semantic segmentation. Next, we learn two different full-waveform material classifiers: a random forest classifier and a temporal convolutional neural network (TCN) classifier. We find that, in some cases, material types can be distinguished, and that the TCN generally performs better across a wider range of materials. However, factors such as angle of incidence, material colour, and material similarity may hinder overall performance.
The popularity of Neural Radiance Fields (NeRFs) for view synthesis has led to a desire for NeRF editing tools. Here, we focus on inpainting regions in a view-consistent and controllable manner. In addition to the typical NeRF inputs and masks delineating the unwanted region in each view, we require only a single inpainted view of the scene, i.e., a reference view. We use monocular depth estimators to back-project the inpainted view to the correct 3D positions. Then, via a novel rendering technique, a bilateral solver can construct view-dependent effects in non-reference views, making the inpainted region appear consistent from any view. For non-reference disoccluded regions, which cannot be supervised by the single reference view, we devise a method based on image inpainters to guide both the geometry and appearance. Our approach shows superior performance to NeRF inpainting baselines, with the additional advantage that a user can control the generated scene via a single inpainted image. Project page: https://ashmrz.github.io/reference-guided-3d
To operate safely and efficiently alongside human workers, collaborative robots (cobots) require the ability to quickly understand the dynamics of manipulated objects. However, traditional methods for estimating the full set of inertial parameters rely on motions that are necessarily fast and unsafe (to achieve a sufficient signal-to-noise ratio). In this work, we take an alternative approach: by combining visual and force-torque measurements, we develop an inertial parameter identification algorithm that requires slow or 'stop-and-go' motions only, and hence is ideally tailored for use around humans. Our technique, called Homogeneous Part Segmentation (HPS), leverages the observation that man-made objects are often composed of distinct, homogeneous parts. We combine a surface-based point clustering method with a volumetric shape segmentation algorithm to quickly produce a part-level segmentation of a manipulated object; the segmented representation is then used by HPS to accurately estimate the object's inertial parameters. To benchmark our algorithm, we create and utilize a novel dataset consisting of realistic meshes, segmented point clouds, and inertial parameters for 20 common workshop tools. Finally, we demonstrate the real-world performance and accuracy of HPS by performing an intricate 'hammer balancing act' autonomously and online with a low-cost collaborative robotic arm. Our code and dataset are open source and freely available.
Correct radar data fusion depends on knowledge of the spatial transform between sensor pairs. Current methods for determining this transform operate by aligning identifiable features in different radar scans, or by relying on measurements from another, more accurate sensor (e.g., a lidar unit). Feature-based alignment requires the sensors to have overlapping fields of view or necessitates the construction of an environment map. Several existing methods require bespoke retroreflective radar targets. These requirements limit both where and how calibration can be performed. In this paper, we take a different approach: instead of attempting to track targets or features, which can be difficult in noisy radar data, we instead rely on ego-velocity estimates from each radar to perform calibration. Our method enables calibration of a subset of the transform parameters, including the yaw and axis of translation between the radar pair, without the need for a shared field of view or for specialized structures in the environment. In general, the yaw and axis of translation are the most important parameters for data fusion, the most likely to vary over time, and the most difficult to calibrate manually. We formulate calibration as a batch optimization problem, prove that the radar-radar system is identifiable, and specify the platform excitation requirements. Through simulations studies and real-world experiments, we establish that our method is more reliable and accurate at estimating the yaw and translation axis than state-of-the-art methods. Finally, we show that the full rigid-body transform can be recovered if relatively coarse information about the rotation rate is available.
Reducing the quantity of annotations required for supervised training is vital when labels are scarce and costly. This reduction is particularly important for semantic segmentation tasks involving 3D datasets, which are often significantly smaller and more challenging to annotate than their image-based counterparts. Self-supervised pre-training on unlabelled data is one way to reduce the amount of manual annotations needed. Previous work has focused on pre-training with point clouds exclusively. While useful, this approach often requires two or more registered views. In the present work, we combine image and point cloud modalities by first learning self-supervised image features and then using these features to train a 3D model. By incorporating image data, which is often included in many 3D datasets, our pre-training method only requires a single scan of a scene and can be applied to cases where localization information is unavailable. We demonstrate that our pre-training approach, despite using single scans, achieves comparable performance to other multi-scan, point cloud-only methods.
Adversarial imitation learning (AIL) has become a popular alternative to supervised imitation learning that reduces the distribution shift suffered by the latter. However, AIL requires effective exploration during an online reinforcement learning phase. In this work, we show that the standard, naive approach to exploration can manifest as a suboptimal local maximum if a policy learned with AIL sufficiently matches the expert distribution without fully learning the desired task. This can be particularly catastrophic for manipulation tasks, where the difference between an expert and a non-expert state-action pair is often subtle. We present Learning from Guided Play (LfGP), a framework in which we leverage expert demonstrations of multiple exploratory, auxiliary tasks in addition to a main task. The addition of these auxiliary tasks forces the agent to explore states and actions that standard AIL may learn to ignore. Additionally, this particular formulation allows for the reusability of expert data between main tasks. Our experimental results in a challenging multitask robotic manipulation domain indicate that LfGP significantly outperforms both AIL and behaviour cloning, while also being more expert sample efficient than these baselines. To explain this performance gap, we provide further analysis of a toy problem that highlights the coupling between a local maximum and poor exploration, and also visualize the differences between the learned models from AIL and LfGP.
Reducing the quantity of annotations required for supervised training is vital when labels are scarce and costly. This reduction is especially important for semantic segmentation tasks involving 3D datasets that are often significantly smaller and more challenging to annotate than their image-based counterparts. Self-supervised pre-training on large unlabelled datasets is one way to reduce the amount of manual annotations needed. Previous work has focused on pre-training with point cloud data exclusively; this approach often requires two or more registered views. In the present work, we combine image and point cloud modalities, by first learning self-supervised image features and then using these features to train a 3D model. By incorporating image data, which is often included in many 3D datasets, our pre-training method only requires a single scan of a scene. We demonstrate that our pre-training approach, despite using single scans, achieves comparable performance to other multi-scan, point cloud-only methods.
Neural Radiance Fields (NeRFs) have emerged as a popular approach for novel view synthesis. While NeRFs are quickly being adapted for a wider set of applications, intuitively editing NeRF scenes is still an open challenge. One important editing task is the removal of unwanted objects from a 3D scene, such that the replaced region is visually plausible and consistent with its context. We refer to this task as 3D inpainting. In 3D, solutions must be both consistent across multiple views and geometrically valid. In this paper, we propose a novel 3D inpainting method that addresses these challenges. Given a small set of posed images and sparse annotations in a single input image, our framework first rapidly obtains a 3D segmentation mask for a target object. Using the mask, a perceptual optimizationbased approach is then introduced that leverages learned 2D image inpainters, distilling their information into 3D space, while ensuring view consistency. We also address the lack of a diverse benchmark for evaluating 3D scene inpainting methods by introducing a dataset comprised of challenging real-world scenes. In particular, our dataset contains views of the same scene with and without a target object, enabling more principled benchmarking of the 3D inpainting task. We first demonstrate the superiority of our approach on multiview segmentation, comparing to NeRFbased methods and 2D segmentation approaches. We then evaluate on the task of 3D inpainting, establishing state-ofthe-art performance against other NeRF manipulation algorithms, as well as a strong 2D image inpainter baseline