Understanding the semantic characteristics of the environment is a key enabler for autonomous robot operation. In this paper, we propose a deep convolutional neural network (DCNN) for semantic segmentation of a LiDAR scan into the classes car, pedestrian and bicyclist. This architecture is based on dense blocks and efficiently utilizes depth separable convolutions to limit the number of parameters while still maintaining the state-of-the-art performance. To make the predictions from the DCNN temporally consistent, we propose a Bayes filter based method. This method uses the predictions from the neural network to recursively estimate the current semantic state of a point in a scan. This recursive estimation uses the knowledge gained from previous scans, thereby making the predictions temporally consistent and robust towards isolated erroneous predictions. We compare the performance of our proposed architecture with other state-of-the-art neural network architectures and report substantial improvement. For the proposed Bayes filter approach, we shows results on various sequences in the KITTI tracking benchmark.

Robust data association is necessary for virtually every SLAM system and finding corresponding points is typically a preprocessing step for scan alignment algorithms. Traditionally, handcrafted feature descriptors were used for these problems but recently learned descriptors have been shown to perform more robustly. In this work, we propose a local feature descriptor for 3D LiDAR scans. The descriptor is learned using a Convolutional Neural Network (CNN). Our proposed architecture consists of a Siamese network for learning a feature descriptor and a metric learning network for matching the descriptors. We also present a method for estimating local surface patches and obtaining ground-truth correspondences. In extensive experiments, we compare our learned feature descriptor with existing 3D local descriptors and report highly competitive results for multiple experiments in terms of matching accuracy and computation time.

Robots are expected to operate autonomously in dynamic environments. Understanding the underlying dynamic characteristics of objects is a key enabler for achieving this goal. In this paper, we propose a method for pointwise semantic classification of 3D LiDAR data into three classes: non-movable, movable and dynamic. We concentrate on understanding these specific semantics because they characterize important information required for an autonomous system. To learn the distinction between movable and non-movable points in the environment, we introduce an approach based on deep neural network and for detecting the dynamic points, we estimate pointwise motion. We propose a Bayes filter framework for combining the learned semantic cues with the motion cues to infer the required semantic classification. In extensive experiments, we compare our approach with other methods on a standard benchmark dataset and report competitive results in comparison to the existing state-of-the-art. Furthermore, we show an improvement in the classification of points by combining the semantic cues retrieved from the neural network with the motion cues.

Robots are expected to operate autonomously in increasingly complex scenarios such as crowded streets or heavy traffic situations. Perceiving the dynamics of moving objects in the environment is crucial for safe and smart navigation and therefore a key enabler for autonomous driving. In this paper we present a novel model-free approach for detecting and tracking dynamic objects in 3D LiDAR scans obtained by a moving sensor. Our method only relies on motion cues and does not require any prior information about the objects. We sequentially detect multiple motions in the scene and segment objects using a Bayesian approach. For robustly tracking objects, we utilize their estimated motion models. We present extensive quantitative results based on publicly available datasets and show that our approach outperforms the state of the art.

The perception of the dynamic aspects of the environment is a highly relevant precondition for the realization of autonomous robot system acting in the real world. In this paper, we propose a novel method for estimating dense rigid scene flow in 3D LiDAR scans. We formulate the problem as an energy minimization problem, where we assume local geometric constancy and incorporate regularization for smooth motion fields. Analyzing the dynamics at point level helps in inferring the fine-grained details of motion. We show results on multiple sequences of the KITTI odometry dataset, where we seamlessly estimate multiple motions pertaining to different dynamic objects. Furthermore, we test our approach on a dataset with pedestrians to show how our method adapts to a case with non-rigid motion. For comparison we use the ground truth from KITTI and show how our method outperforms different ICP-based methods.

This paper presents a novel exploration strategy for coordinated exploration between unmanned ground vehicles (UGV) and micro-air vehicles (MAV). The exploration is modeled as an Integer Programming (IP) optimization problem and the allocation of the vehicles(agents) to frontier locations is modeled using binary variables. The formulation is also studied for distributed system, where agents are divided into multiple teams using graph partitioning. Optimization seamlessly integrates several practical constraints that arise in exploration between such heterogeneous agents and provides an elegant solution for assigning task to agents. We have also presented comparison with previous methods based on distance traversed and computational time to signify advantages of presented method. We also show practical realization of such an exploration where an UGV-MAV team efficiently builds a map of an indoor environment.