Search Results

Traditional ground vehicle architectures comprise of a chassis connected via passive, semi-active, or active suspension systems to multiple ground wheels. Current design-optimizations of vehicle architectures for on-road applications have diminished their mobility and maneuverability in off-road settings. Autonomous Ground Vehicles (AGV) traversing off-road environments face numerous challenges concerning terrain roughness, soil hardness, uneven obstacle-filled terrain, and varying traction conditions. Numerous Active Articulated-Wheeled (AAW) vehicle architectures have emerged to permit AGVs to adapt to variable terrain conditions in various off-road application arenas (off-road, construction, mining, and space robotics). However, a comprehensive framework of AAW platforms for exploring various facets of system architecture/design, analysis (kinematics/dynamics), and control (motions/forces) remains challenging.

Simultaneous Localization and Mapping (SLAM) algorithms are extensively utilized within the field of autonomous navigation. In particular, numerous open-source Robot Operating System (ROS) based SLAM solutions, such as Gmapping, Hector, Cartographer etc., have simplified deployments in application. However, establishing the accuracy and precision of these ‘out-of-the-box’ SLAM algorithms is necessary for improving the accuracy and precision of further applications such as planning, navigation, controls. Existing benchmarking literature largely focused on validating SLAM algorithms based upon the quality of the generated maps. In this paper, however, we focus on examining the localization accuracy of existing 2-dimensional LiDAR based indoor SLAM algorithms. The fidelity of these implementations is compared against the OptiTrack motion capture system which is capable of tracking moving objects at sub-millimeter level precision.

Collaborative robots are more and more used in smart manufacturing because of their capability to work beside and collaborate with human workers. With the deployment of these robots, manufacturing tasks are more inclined to be accomplished by multiple humans and multiple robots (MH-MR) through teaming effort. In such MH-MR collaboration scenarios, the task distribution among the multiple humans and multiple robots is very critical to efficiency. It is also more challenging due to the heterogeneity of different agents. Existing approaches in task distribution among multiple agents mostly consider humans with assumed or known capabilities. However human capabilities are always changing due to various factors, which may lead to suboptimal efficiency. Although some researches have studied several human factors in manufacturing and applied them to adjust the robot task and behaviors.

Integrated modeling of vehicle, tire and terrain is a fundamental challenge to be addressed for off-road autonomous navigation. The complexities arise due to lack of tools and techniques to predict the continuously varying terrain and environmental conditions and the resultant non-linearities. The solution to this challenge can now be found in the plethora of data driven modeling and control techniques that have gained traction in the last decade. Data driven modeling and control techniques rely on the system’s repeated interaction with the environment to generate a lot of data and then use a function approximator to fit a model for the physical system with the data. Getting good quality and quantity of data may involve extensive experimentation with the physical system impacting developer’s resource. The process is computationally expensive, and the overhead time required is high.

Modeling uncertainties pose a significant challenge in the development and deployment of model-based vehicle control systems. Most model- based automotive control systems require the use of a well estimated vehicle dynamics prediction model. The ability of first principles-based models to represent vehicle behavior becomes limited under complex scenarios due to underlying rigid physical assumptions. Additionally, the increasing complexity of these models to meet ever-increasing fidelity requirements presents challenges for obtaining analytical solutions as well as control design. Alternatively, deterministic data driven techniques including but not limited to deep neural networks, polynomial regression, Sparse Identification of Nonlinear Dynamics (SINDy) have been deployed for vehicle dynamics system identification and prediction.

Capacitance sensing is the technology that detects the presence of nearby objects by measuring the change in capacitance. A change in capacitance is triggered either by a change in dielectric constant, area of overlap or distance of separation between the electrodes of the capacitor. It is a technology that finds wide use in applications such as touch screens, proximity sensing etc. Drawing motivation from such applications, this paper investigates how capacitive sensing can be employed to detect the presence and posture of occupants inside vehicles. Compared to existing solutions, the proposed approach is low-cost, easy to deploy and highly efficient. The sensing system consists of a capacitance-sensing mat that is embedded with copper foils and an associated sensing circuitry. Inside the mat the foils are arranged in rows and columns to form several touch-nodes across the surface of the mat.

Passenger comfort is a critical factor in user acceptance of autonomous vehicles (AVs). Despite existing methods for passenger comfort assessment, new approaches to assessing passenger comfort in AVs may be valuable to the automotive industry. In this paper, continuous pressing-based and discrete smartphone-based approaches for comfort assessment were designed and implemented in a user study. Participants used the two approaches to evaluate their comfort levels in an experimental study based on a high-fidelity autonomous driving simulator. Performances of the two approaches in assessing comfort levels were analyzed and compared. In general, the discrete approach showed better measurement repeatability and lower measurement bias than the continuous approach. The performance gap of the continuous approach could be reduced with proper post-processing measures. Discussions on the potential uses of the approaches were also raised.

Enabling self-driving vehicles to efficiently and autonomously navigate through an obstacle-filled environment remains a topic of significant contemporary research interest. Motion-planning frameworks, encapsulating both path- and trajectory-planning, have played a dominant role in realizing the deployment of a “sense-think-act” intelligence for autonomous vehicles. However, verification and validation of such intelligence on actual self-driving autonomous vehicles has been limited. Simulation-based verification and validation has the advantage of permitting diverse scenario-based testing and comprehensive “what-if” analyses - but is ultimately limited by the simulation fidelity and realism. In contrast, testing on full-scale real-world systems is constrained by the usual challenges of time, space, and cost engendered in reproducing diverse scenarios in practice.

Learning only from a small set of examples remains a huge challenge in machine learning. Despite recent breakthroughs in the applications of neural networks, the applicability of these techniques has been limited by the requirement for large amounts of training data. What’s more, the standard supervised machine learning method does not provide a satisfactory solution for learning new concepts from little data. However, the ability to learn enough information from few samples has been demonstrated in humans. This suggests that humans may make use of prior knowledge of a previously learned model when learning new ones on a small amount of training examples. In the area of autonomous driving, the model learns to drive the vehicle with training data from humans, and most machine learning based control algorithms require training on very large datasets. Collecting and constructing training data set takes a huge amount of time and needs specific knowledge to gather relevant information.

Over the years, the complexity of autonomous vehicle development (and concurrently the verification and validation) has grown tremendously in terms of component-, subsystem- and system-level interactions between autonomy and the human users. Simulation-based testing holds significant promise in helping to identify both problematic interactions between component-, subsystem-, and system-levels as well as overcoming delays typically introduced by the default full-scale on-road testing. Software in Loop (SiL) simulation is utilized as an intermediate step towards software deployment for autonomous vehicles (AV) to make them reliable. SiL efforts can help reduce the resources required for successful deployment by helping to validate the software for millions of road miles. A key enabler for accelerating SiL processes is the ability to use Simulation as a Service (SaaS) rather than just isolated instances of software.

Recent trends in autonomy have emphasized end-to-end deep-learning-based methods that have shown a lot of promise in overcoming the requirements and limitations of feature-engineering. However, while promising, the black-box nature of deep-learning frameworks now exacerbates the need for testing with end-to-end deployments. Further, as exemplars of systems-of-systems, autonomous vehicles (AVs) engender numerous interconnected component-, subsystem and system-level interactions. The ensuing complexity creates challenges for verification and validation at the various component, subsystem- and system-levels as well as end-to-end testing. While simulation-based testing is one promising avenue, oftentimes the lack of adequate fidelity of AV and environmental modeling limits the generalizability. In contrast, full-scale AV testing presents the usual limitations of time-, space-, and cost.

This paper presents an approach for performing software in the loop testing of autonomous vehicle software developed in the Autoware framework. Autoware is an open source software for autonomous driving that includes modules such as localization, detection, prediction, planning and control [8]. Multitudes of autonomous driving frameworks exist today, each having its own pros and cons. Often, MATLAB-Simulink is used for rapid prototyping, system modeling and testing, specifically for the lower-level vehicle dynamics and powertrain control features. For the autonomous software, the Robotic Operating System (ROS) is more commonly used for integrating distributed software components so that they can easily share information through a publish and subscribe paradigm. Thorough testing and evaluation of such complex, distributed software, implemented on a physical vehicle poses significant challenges in terms of safety, time, and cost, especially when considering rare edge cases.

In this paper, we present a framework for the robot to learn how to place objects to a workpiece by learning from humans in smart manufacturing. In the proposed framework, the rational scene dictionary (RSD) corresponding to the keyframes of task (KFT) are used to identify the general object-action-location relationships. The Generalized Voronoi Diagrams (GVD) based contour is used to determine the relative position and orientation between the object and the corresponding workpiece at the final state. In the learning phase, we keep tracking the image segments in the human demonstration. For the moment when a spatial relation of some segments are changed in a discontinuous way, the state changes are recorded by the RSD. KFT is abstracted after traversing and searching in RSD, while the relative position and orientation of the object and the corresponding mount are presented by GVD-based contours for the keyframes.

Over the last decade, tremendous amount of research and progress has been made towards developing smart technologies for autonomous vehicles such as adaptive cruise control, lane keeping assist, lane following algorithms, and decision-making algorithms. One of the fundamental objectives for the development of such technologies is to enable autonomous vehicles with the capability to avoid obstacles and maintain safety. Automobiles are real-world dynamical systems - possessing inertia, operating at varying speeds, with finite accelerations/decelerations during operations. Deployment of autonomy in vehicles increases in complexity multi-fold especially when high DOF vehicle models need to be considered for robust control. Model Predictive Control (MPC) is a powerful tool that is used extensively to control the behavior of complex, dynamic systems. As a model-based approach, the fidelity of the model and selection of model-parameters plays a role in ultimate performance.

Fully autonomous vehicles have the potential to fundamentally transform the future transportation system. While previous research has examined individuals’ perceptions towards fully autonomous vehicles, a complete understanding of attitudes and opinions across the lifespan is unknown. Therefore, individuals’ awareness, acceptance, and preferences towards autonomous vehicles were obtained from 75 participants through interviews with three diverse groups of participants: 20 automotive engineering graduate students who were building an autonomous concept vehicle, 21 non-technical adults, and 34 senior citizens. The results showed that regardless of age, an individual’s readiness to ride in a fully autonomous vehicle and the vehicle’s requirements were influenced by the users’ understanding of autonomous vehicles.

It’s important to predict human actions in the industry assembly process. Foreseeing future actions before they happened is an essential part for flexible human-robot collaboration and crucial to safety issues. Vision-based human action prediction from videos provides intuitive and adequate knowledge for many complex applications. This problem can be interpreted as deducing the next action of people from a short video clip. The history information needs to be considered to learn these relations among time steps for predicting the future steps. However, it is difficult to extract the history information and use it to infer the future situation with traditional methods. In this scenario, a model is needed to handle the spatial and temporal details stored in the past human motions and construct the future action based on limited accessible human demonstrations.

Range anxiety and lack of adequate access to fast charging are proving to be important impediments to electric vehicle (EV) adoption. While many techniques to fast charging EV batteries (model-based & model-free) have been developed, they have focused on a single Lithium-ion cell. Extensions to battery packs are scarce, often considering simplified architectures (e.g., series-connected) for ease of modeling. Computational considerations have also restricted fast-charging simulations to small battery packs, e.g., four cells (for both series and parallel connected cells). Hence, in this paper, we pursue a model-free approach based on reinforcement learning (RL) to fast charge a large battery pack (comprising 444 cells). Each cell is characterized by an equivalent circuit model coupled with a second-order lumped thermal model to simulate the battery behavior. After training the underlying RL, the developed model will be straightforward to implement with low computational complexity.

Autonomous driving technologies can provide better safety, comfort and efficiency for future transportation systems. Most research in this area has mainly been focused on developing sensing and control approaches to achieve various autonomous driving functions. Very little of this research, however, has studied how to efficiently handle sensing exceptions. A simple exception measured by any of the sensors may lead to failures in autonomous driving functions. The autonomous vehicles are then supposed to be sent back to manufacturers for repair, which takes both time and money. This paper introduces an efficient approach to make human drivers able to online teach autonomous vehicles to drive under sensing exceptions. A human-vehicle teaching-and-learning framework for autonomous driving is proposed and the human teaching and vehicle learning processes for handling sensing exceptions in autonomous vehicles are designed in detail.

Improving passenger safety inside vehicle cabins requires continuously monitoring vehicle seat occupancy statuses. Monitoring a vehicle seat’s occupancy status includes detecting if the seat is occupied and classifying the seat’s occupancy type. This paper introduces an innovative non-intrusive technique that employs capacitive sensing and an occupancy classifier to monitor a vehicle seat’s occupancy status. Capacitive sensing is facilitated by a meticulously constructed capacitance-sensing mat that easily integrates with any vehicle seat. When a passenger or an inanimate object occupies a vehicle seat equipped with the mat, they will induce variations in the mat’s internal capacitances. The variations are, in turn, represented pictorially as grayscale capacitance-sensing images (CSI), which yield the feature vectors the classifier requires to classify the seat’s occupancy type.

Artificial intelligence (AI) enhanced control system deployments are emerging as a viable substitute to more traditional control system. In particular, deep learning techniques offer an alternate approach to tune the ever increasing sets of control system parameters to extract performance. However, the systematic verification and validation (to establish the reliability and robustness) of deep learning based controllers in actual deployments remains a challenge. This is exacerbated by the need to evaluate and optimize control systems embedded within an operational environment (with its own sets of additional unknown or uncertain parameters). Existing literature comparisons of deep learning against traditional controllers, where they may exist, do not offer structured approaches to comparative performance evaluation and improvement. It is also crucial to develop a standardized controlled test environment within which various controllers are evaluated against a common metric.