Learning from Local Experience: Informed Sampling Distributions for High Dimensional Motion Planning

This paper presents a sampling-based motion planning framework that leverages the geometry of obstacles in a workspace as well as prior experiences from motion planning problems. Previous studies have demonstrated the benefits of utilizing prior solutions to motion planning problems for improving planning efficiency. However, particularly for high-dimensional systems, achieving high performance across randomized environments remains a technical challenge for experience-based approaches due to the substantial variance between each query. To address this challenge, we propose a novel approach that involves decoupling the problem into subproblems through algorithmic workspace decomposition and graph search. Additionally, we capitalize on prior experience within each subproblem. This approach effectively reduces the variance across different problems, leading to improved performance for experience-based planners. To validate the effectiveness of our framework, we conduct experiments using 2D and 6D robotic systems. The experimental results demonstrate that our framework outperforms existing algorithms in terms of planning time and cost.

Residual Q-Learning: Offline and Online Policy Customization without Value

Imitation Learning (IL) is a widely used framework for learning imitative behavior from demonstrations. It is especially appealing for solving complex real-world tasks where handcrafting reward function is difficult, or when the goal is to mimic human expert behavior. However, the learned imitative policy can only follow the behavior in the demonstration. When applying the imitative policy, we may need to customize the policy behavior to meet different requirements coming from diverse downstream tasks. Meanwhile, we still want the customized policy to maintain its imitative nature. To this end, we formulate a new problem setting called policy customization. It defines the learning task as training a policy that inherits the characteristics of the prior policy while satisfying some additional requirements imposed by a target downstream task. We propose a novel and principled approach to interpret and determine the trade-off between the two task objectives. Specifically, we formulate the customization problem as a Markov Decision Process (MDP) with a reward function that combines 1) the inherent reward of the demonstration; and 2) the add-on reward specified by the downstream task. We propose a novel framework, Residual Q-learning, which can solve the formulated MDP by leveraging the prior policy without knowing the inherent reward or value function of the prior policy. We derive a family of residual Q-learning algorithms that can realize offline and online policy customization, and show that the proposed algorithms can effectively accomplish policy customization tasks in various environments.

Efficient Multi-Task and Transfer Reinforcement Learning with Parameter-Compositional Framework

In this work, we investigate the potential of improving multi-task training and also leveraging it for transferring in the reinforcement learning setting. We identify several challenges towards this goal and propose a transferring approach with a parameter-compositional formulation. We investigate ways to improve the training of multi-task reinforcement learning which serves as the foundation for transferring. Then we conduct a number of transferring experiments on various manipulation tasks. Experimental results demonstrate that the proposed approach can have improved performance in the multi-task training stage, and further show effective transferring in terms of both sample efficiency and performance.

Double-Iterative Gaussian Process Regression for Modeling Error Compensation in Autonomous Racing

Autonomous racing control is a challenging research problem as vehicles are pushed to their limits of handling to achieve an optimal lap time; therefore, vehicles exhibit highly nonlinear and complex dynamics. Difficult-to-model effects, such as drifting, aerodynamics, chassis weight transfer, and suspension can lead to infeasible and suboptimal trajectories. While offline planning allows optimizing a full reference trajectory for the minimum lap time objective, such modeling discrepancies are particularly detrimental when using offline planning, as planning model errors compound with controller modeling errors. Gaussian Process Regression (GPR) can compensate for modeling errors. However, previous works primarily focus on modeling error in real-time control without consideration for how the model used in offline planning can affect the overall performance. In this work, we propose a double-GPR error compensation algorithm to reduce model uncertainties; specifically, we compensate both the planner's model and controller's model with two respective GPR-based error compensation functions. Furthermore, we design an iterative framework to re-collect error-rich data using the racing control system. We test our method in the high-fidelity racing simulator Gran Turismo Sport (GTS); we find that our iterative, double-GPR compensation functions improve racing performance and iteration stability in comparison to a single compensation function applied merely for real-time control.

Learning Generalizable Pivoting Skills

The skill of pivoting an object with a robotic system is challenging for the external forces that act on the system, mainly given by contact interaction. The complexity increases when the same skills are required to generalize across different objects. This paper proposes a framework for learning robust and generalizable pivoting skills, which consists of three steps. First, we learn a pivoting policy on an ``unitary'' object using Reinforcement Learning (RL). Then, we obtain the object's feature space by supervised learning to encode the kinematic properties of arbitrary objects. Finally, to adapt the unitary policy to multiple objects, we learn data-driven projections based on the object features to adjust the state and action space of the new pivoting task. The proposed approach is entirely trained in simulation. It requires only one depth image of the object and can zero-shot transfer to real-world objects. We demonstrate robustness to sim-to-real transfer and generalization to multiple objects.

Quadric Representations for LiDAR Odometry, Mapping and Localization

Current LiDAR odometry, mapping and localization methods leverage point-wise representations of 3D scenes and achieve high accuracy in autonomous driving tasks. However, the space-inefficiency of methods that use point-wise representations limits their development and usage in practical applications. In particular, scan-submap matching and global map representation methods are restricted by the inefficiency of nearest neighbor searching (NNS) for large-volume point clouds. To improve space-time efficiency, we propose a novel method of describing scenes using quadric surfaces, which are far more compact representations of 3D objects than conventional point clouds. In contrast to point cloud-based methods, our quadric representation-based method decomposes a 3D scene into a collection of sparse quadric patches, which improves storage efficiency and avoids the slow point-wise NNS process. Our method first segments a given point cloud into patches and fits each of them to a quadric implicit function. Each function is then coupled with other geometric descriptors of the patch, such as its center position and covariance matrix. Collectively, these patch representations fully describe a 3D scene, which can be used in place of the original point cloud and employed in LiDAR odometry, mapping and localization algorithms. We further design a novel incremental growing method for quadric representations, which eliminates the need to repeatedly re-fit quadric surfaces from the original point cloud. Extensive odometry, mapping and localization experiments on large-volume point clouds in the KITTI and UrbanLoco datasets demonstrate that our method maintains low latency and memory utility while achieving competitive, and even superior, accuracy.

SparseFusion: Fusing Multi-Modal Sparse Representations for Multi-Sensor 3D Object Detection

By identifying four important components of existing LiDAR-camera 3D object detection methods (LiDAR and camera candidates, transformation, and fusion outputs), we observe that all existing methods either find dense candidates or yield dense representations of scenes. However, given that objects occupy only a small part of a scene, finding dense candidates and generating dense representations is noisy and inefficient. We propose SparseFusion, a novel multi-sensor 3D detection method that exclusively uses sparse candidates and sparse representations. Specifically, SparseFusion utilizes the outputs of parallel detectors in the LiDAR and camera modalities as sparse candidates for fusion. We transform the camera candidates into the LiDAR coordinate space by disentangling the object representations. Then, we can fuse the multi-modality candidates in a unified 3D space by a lightweight self-attention module. To mitigate negative transfer between modalities, we propose novel semantic and geometric cross-modality transfer modules that are applied prior to the modality-specific detectors. SparseFusion achieves state-of-the-art performance on the nuScenes benchmark while also running at the fastest speed, even outperforming methods with stronger backbones. We perform extensive experiments to demonstrate the effectiveness and efficiency of our modules and overall method pipeline. Our code will be made publicly available at https://github.com/yichen928/SparseFusion.

Open-Vocabulary Point-Cloud Object Detection without 3D Annotation

The goal of open-vocabulary detection is to identify novel objects based on arbitrary textual descriptions. In this paper, we address open-vocabulary 3D point-cloud detection by a dividing-and-conquering strategy, which involves: 1) developing a point-cloud detector that can learn a general representation for localizing various objects, and 2) connecting textual and point-cloud representations to enable the detector to classify novel object categories based on text prompting. Specifically, we resort to rich image pre-trained models, by which the point-cloud detector learns localizing objects under the supervision of predicted 2D bounding boxes from 2D pre-trained detectors. Moreover, we propose a novel de-biased triplet cross-modal contrastive learning to connect the modalities of image, point-cloud and text, thereby enabling the point-cloud detector to benefit from vision-language pre-trained models,i.e.,CLIP. The novel use of image and vision-language pre-trained models for point-cloud detectors allows for open-vocabulary 3D object detection without the need for 3D annotations. Experiments demonstrate that the proposed method improves at least 3.03 points and 7.47 points over a wide range of baselines on the ScanNet and SUN RGB-D datasets, respectively. Furthermore, we provide a comprehensive analysis to explain why our approach works.

A Simple Approach for General Task-Oriented Picking using Placing constraints

Pick-and-place is an important manipulation task in domestic or manufacturing applications. There exist many works focusing on grasp detection with high picking success rate but lacking consideration of downstream manipulation tasks (e.g., placing). Although some research works proposed methods to incorporate task conditions into grasp selection, most of them are data-driven and are therefore hard to adapt to arbitrary operating environments. Observing this challenge, we propose a general task-oriented pick-place framework that treats the target task and operating environment as placing constraints into grasping optimization. Combined with existing grasp detectors, our framework is able to generate feasible grasps for different downstream tasks and adapt to environmental changes without time-consuming re-training processes. Moreover, the framework can accept different definitions of placing constraints, so it is easy to integrate with other modules. Experiments in the simulator and real-world on multiple pick-place tasks are conducted to evaluate the performance of our framework. The result shows that our framework achieves a high and robust task success rate on a wide variety of the pick-place tasks.

Active Finetuning: Exploiting Annotation Budget in the Pretraining-Finetuning Paradigm

Given the large-scale data and the high annotation cost, pretraining-finetuning becomes a popular paradigm in multiple computer vision tasks. Previous research has covered both the unsupervised pretraining and supervised finetuning in this paradigm, while little attention is paid to exploiting the annotation budget for finetuning. To fill in this gap, we formally define this new active finetuning task focusing on the selection of samples for annotation in the pretraining-finetuning paradigm. We propose a novel method called ActiveFT for active finetuning task to select a subset of data distributing similarly with the entire unlabeled pool and maintaining enough diversity by optimizing a parametric model in the continuous space. We prove that the Earth Mover's distance between the distributions of the selected subset and the entire data pool is also reduced in this process. Extensive experiments show the leading performance and high efficiency of ActiveFT superior to baselines on both image classification and semantic segmentation. Our code is released at https://github.com/yichen928/ActiveFT.