Radar odometry has been gaining attention in the last decade. It stands as one of the best solutions for robotic state estimation in unfavorable conditions; conditions where other interoceptive and exteroceptive sensors may fall short. Radars are widely adopted, resilient to weather and illumination, and provide Doppler information which make them very attractive for such tasks. This article presents an extensive survey of the latest work on ground-based radar odometry for autonomous robots. It covers technologies, datasets, metrics, and approaches that have been developed in the last decade in addition to in-depth analysis and categorization of the various methods and techniques applied to tackle this problem. This article concludes with challenges and future recommendations to advance the field of radar odometry making it a great starting point for newcomers and a valuable reference for experienced researchers.

Autonomous vehicles rely on perceiving the road environment through a perception pipeline fed by a variety of sensor modalities, including camera, lidar, radar, infrared, gated camera, and ultrasonic. The vehicle computer must perform sensor fusion reliably in various sensor signals degrading environments, which can be performed parametrically or using machine learning. We study sensor fusion of a forward-facing camera and a lidar sensor using convolutional neural networks (CNNs), for drivable area detection in winter driving, where the roads are partially covered with snow, tire tracks obscuring lane markers. Seven fusion models are developed and evaluated for their ability to classify pixels into two classes: drivable and nondrivable. A total of seven models are designed and tested, including camera only, lidar only, early fusion, (three types of) intermediate fusion, and late fusion. The models have accuracies between 84% and 89%, with runtimes between 23 and 61 ms. To select the best model from this group, we introduce a unique metric, named normalized accuracy runtime (NAR) score. A network in a group has a higher NAR score, the more accurate it is, and the shorter its runtime is. The models are evaluated on a winter driving DENSE dataset. Camera and lidar fog noise are synthesized to examine model robustness.

In this work, a design of a gripper for the underwater OpenROV vehicle is presented. OpenROV is an open-source underwater vehicle design for remote underwater exploration. It can enable systems of underwater internet of things and real-time monitoring. Mechanical aspects of the presented gripper design are discussed including actuation, motion transmission, kinematics and general arrangement, which resembles a delta robot. The Denavit-Hartenberg (DH) notation will be employed to define reference frames on one of the fingers in order to build transformation matrices and the forward kinematics matrix. The results from the forward kinematics are used to define the workspace that can be covered by each finger. The maximum force from the fingertip is estimated using Newton-Euler equations. Finally, the transfer function and the mass moment of inertia of the second link in the finger, that is, the fingertip is calculated for control simulations. A control stability analysis is provided and shows a stable system.