Robotics (cs.RO)

Abstract: This work presents the mechanical design and control of a novel small-size and lightweight Micro Aerial Vehicle (MAV) for aerial manipulation. To our knowledge, with a total take-off mass of only 2.0 kg, the proposed system is the most lightweight Aerial Manipulator (AM) that has 8-DOF independently controllable: 5 for the aerial platform and 3 for the articulated arm. We designed the robot to be fully-actuated in the body forward direction. This allows independent pitching and instantaneous force generation, improving the platform's performance during physical interaction. The robotic arm is an origami delta manipulator driven by three servomotors, enabling active motion compensation at the end-effector. Its composite multimaterial links help reduce the weight, while their flexibility allow for compliant aerial interaction with the environment. In particular, the arm's stiffness can be changed according to its configuration. We provide an in depth discussion of the system design and characterize the stiffness of the delta arm. A control architecture to deal with the platform's overactuation while exploiting the delta arm is presented. Its capabilities are experimentally illustrated both in free flight and physical interaction, highlighting advantages and disadvantages of the origami's folding mechanism.

Abstract: Recent advancements in legged locomotion research have made legged robots a preferred choice for navigating challenging terrains when compared to their wheeled counterparts. This paper presents a novel locomotion policy, trained using Deep Reinforcement Learning, for a quadrupedal robot equipped with an additional prismatic joint between the knee and foot of each leg. The training is performed in NVIDIA Isaac Gym simulation environment. Our study investigates the impact of these joints on maintaining the quadruped's desired height and following commanded velocities while traversing challenging terrains. We provide comparison results, based on a Cost of Transport (CoT) metric, between quadrupeds with and without prismatic joints. The learned policy is evaluated on a set of challenging terrains using the CoT metric in simulation. Our results demonstrate that the added degrees of actuation offer the locomotion policy more flexibility to use the extra joints to traverse terrains that would be deemed infeasible or prohibitively expensive for the conventional quadrupedal design, resulting in significantly improved efficiency.

Abstract: This document describes the architecture and algorithms of a high fidelity fixed wing flight simulator intended to test and validate novel guidance, navigation, and control (GNC) algorithms for autonomous aircraft. It aims to replicate the influence of as many factors as possible on the aircraft performances, the Earth model, the physics of flight and the associated equations of motion, and in particular the behavior of the onboard sensors, limiting the assumptions to the bare minimum, and including multiple relatively minor effects not usually considered in simulation that may play a role in the GNC algorithms not performing as intended. The author releases the flight simulator C ++ implementation as open-source software. The simulator modular design enables the replacement of the standard GNC algorithms with the objective of evaluating their performances when subject to specific missions and meteorological conditions (atmospheric properties, wind field, air turbulence). The testing and evaluation is performed by means of Monte Carlo simulations, as most simulation modules (such as the aircraft mission, the meteorological conditions, the errors introduced by the sensors, and the initial conditions) are defined stochastically and hence vary in a pseudo-random way from one execution to the next according to certain user-defined input parameters, ensuring that the results are valid for a wide range of conditions. In addition to modeling the outputs of all sensors usually present onboard a fixed wing platform, such as accelerometers, gyroscopes, magnetometers, Pitot tube, air vanes, and a Global Navigation Satellite System (GNCC) receiver, the simulator is also capable of generating realistic images of the Earth surface that resemble what an onboard camera would record if following the resulting trajectory, enabling the use and evaluation of visual and visual inertial navigation systems.

Abstract: High Definition (HD) maps are necessary for many applications of automated driving (AD), but their manual creation and maintenance is very costly. Vehicle fleet data from series production vehicles can be used to automatically generate HD maps, but the data is often incomplete and noisy. We propose a system for the generation of HD maps from vehicle fleet data, which is tolerant to missing or misclassified detections and can handle drives with multiple routes, generating a single complete map, model-free and without prior reference lines. Using randomly selected drives as pivot drives, a step-wise lateral sampling of detections is performed. These sampled points are then clustered and aligned using Expectation Maximization (EM), estimating a lateral offset for each drive to compensate localization errors. The clustered points are replaced with the maxima of their probability density function (PDF) and connected to form polylines using a modified rectangular linear assignment algorithm. The data from vehicles on varying routes is then fused into a hierarchical singular map graph. The proposed approach achieves an average accuracy below 0.5 meters compared to a hand annotated ground truth map, as well as correctly resolving lane splits and merges, proving the feasibility of the use of vehicle fleet data for the generation of highway HD maps.

Abstract: The architecture of a robotics software framework tremendously influences the effort and time it takes for end users to test new concepts in a simulation environment and to control real hardware. Many years of activity in the field allowed us to sort out crucial requirements for a framework tailored for robotics: modularity and extensibility, source code reusability, feature richness, and user-friendliness. We implemented these requirements and collected best practices in Locosim, a cross-platform framework for simulation and real hardware. In this paper, we describe the architecture of Locosim and illustrate some use cases that show its potential.

Abstract: Modeling dynamics is often the first step to making a vehicle autonomous. While on-road autonomous vehicles have been extensively studied, off-road vehicles pose many challenging modeling problems. An off-road vehicle encounters highly complex and difficult-to-model terrain/vehicle interactions, as well as having complex vehicle dynamics of its own. These complexities can create challenges for effective high-speed control and planning. In this paper, we introduce a framework for multistep dynamics prediction that explicitly handles the accumulation of modeling error and remains scalable for sampling-based controllers. Our method uses a specially-initialized Long Short-Term Memory (LSTM) over a limited time horizon as the learned component in a hybrid model to predict the dynamics of a 4-person seating all-terrain vehicle (Polaris S4 1000 RZR) in two distinct environments. By only having the LSTM predict over a fixed time horizon, we negate the need for long term stability that is often a challenge when training recurrent neural networks. Our framework is flexible as it only requires odometry information for labels. Through extensive experimentation, we show that our method is able to predict millions of possible trajectories in real-time, with a time horizon of five seconds in challenging off road driving scenarios.

Abstract: This paper investigated the distributed leader follower formation control problem for multiple differentially driven mobile robots. A distributed estimator is first introduced and it only requires the state information from each follower itself and its neighbors. Then, we propose a bioinspired neural dynamic based backstepping and sliding mode control hybrid formation control method with proof of its stability. The proposed control strategy resolves the impractical speed jump issue that exists in the conventional backstepping design. Additionally, considering the system and measurement noises, the proposed control strategy not only removes the chattering issue existing in the conventional sliding mode control but also provides smooth control input with extra robustness. After that, an adaptive sliding innovation filter is integrated with the proposed control to provide accurate state estimates that are robust to modeling uncertainties. Finally, we performed multiple simulations to demonstrate the efficiency and effectiveness of the proposed formation control strategy.