Tulga Ersal

Safe navigation in dynamic and uncertain environments often relies on accurate estimation of, or assumptions about, the true underlying uncertainty. However, accurately characterizing the true uncertainty distribution is often difficult due to limited data or imperfect information. An incorrect understanding of the uncertainty and its associated risk may lead to dangerous decisions even under high levels of risk aversion. To address this issue, we propose a risk-aware model predictive control (RA-MPC) framework that incorporates prediction sets to guarantee risk control below a user-specified threshold without requiring assumptions about the underlying uncertainty distribution. To generate the prediction sets, we develop a distribution-free risk quantification framework that extends conformal risk control (CRC) to general spectral risk measures. We then show that incorporating the prediction sets into the MPC framework provides statistical safety guarantees in terms of spectral risk constraint satisfaction even under uncertainty misspecification. We validate the proposed framework in simulated vehicle obstacle avoidance scenarios, demonstrating improved safety and reduced solve time compared to a baseline RA-MPC framework.

Safe navigation in dynamic and uncertain environments often relies on accurate estimation of, or assumptions about, the true underlying uncertainty. However, accurately characterizing the true uncertainty distribution is often difficult due to limited data or imperfect information. An incorrect understanding of the uncertainty and its associated risk may lead to dangerous decisions even under high levels of risk aversion. To address this issue, we propose a risk-aware model predictive control (RA-MPC) framework that incorporates prediction sets to guarantee risk control below a user-specified threshold without requiring assumptions about the underlying uncertainty distribution. To generate the prediction sets, we develop a distribution-free risk quantification framework that extends conformal risk control (CRC) to general spectral risk measures. We then show that incorporating the prediction sets into the MPC framework provides statistical safety guarantees in terms of spectral risk constraint satisfaction even under uncertainty misspecification. We validate the proposed framework in simulated vehicle obstacle avoidance scenarios, demonstrating improved safety and reduced solve time compared to a baseline RA-MPC framework.

Prior work showed that a three-step serial implementation of the two-point visual steering model in ACT-R can predict the upper bound of human path-keeping performance, but it is unknown whether the cognitive architecture or the number of processing steps drives this ability. This study compares four implementations: no cognitive architecture, ACT-R, and two variants of QN-MHP, each parameterised to minimise average path-keeping error without fitting to human data. Validation against humans reveals that implementations lacking a cognitive architecture or separate processing of the two visual points exceed human performance, failing to predict realistic upper bounds. Two-step serial processing within a cognitive architecture is necessary and sufficient to retain the predictive capabilities for an upper bound of human steering performance.

A novel local trajectory planner, capable of controlling an autonomous off-road vehicle on rugged terrain at high-speed is presented. Autonomous vehicles are currently unable to safely operate off-road at high-speed, as current approaches either fail to predict and mitigate rollovers induced by rough terrain or are not real-time feasible. To address this challenge, a novel model predictive control (MPC) formulation is developed for local trajectory planning. A new dynamics model for off-road vehicles on rough, non-planar terrain is derived and used for prediction. Extreme mobility, including tire liftoff without rollover, is safely enabled through a new energy-based constraint. The formulation is analytically shown to mitigate rollover types ignored by many state-of-the-art methods, and real-time feasibility is achieved through parallelized GPGPU computation. The planner's ability to provide safe, extreme trajectories is studied through both simulated trials and full-scale physical experiments. The results demonstrate fewer rollovers and more successes compared to a state-of-the-art baseline across several challenging scenarios that push the vehicle to its mobility limits.

A novel topology-aware model predictive control (MPC) framework is presented for navigating off-road wheeled vehicles with extreme mobility on 3-D terrains. Prior studies have oversimplified the problem by either treating it solely as a path-planning problem or disregarding the terrain topology. To bridge this gap, this article presents a one-layer model predictive planning and control framework that considers terrain topology for off-road vehicles at operationally relevant speeds. The algorithm is tested in various simulated scenarios with comparisons to benchmarks. The proposed algorithm demonstrates superior performance by capturing the dynamic constraints, while the benchmarks either fail to complete tasks or exhibit inferior performance due to inaccurate representation of the dynamic limitations of the vehicle. Furthermore, the algorithm is tested for robustness to demonstrate its ability to handle uncertainties in terrain topology.

This paper presents a novel envelope based model predictive control (MPC) framework designed to enable autonomous vehicles to handle high performance driving across a wide range of scenarios without a predefined reference. In high performance autonomous driving, safe operation at the vehicle's dynamic limits requires a real time planning and control framework capable of accounting for key vehicle dynamics and environmental constraints when following a predefined reference trajectory is suboptimal or even infeasible. State of the art planning and control frameworks, however, are predominantly reference based, which limits their performance in such situations. To address this gap, this work first introduces a computationally efficient vehicle dynamics model tailored for optimization based control and a continuously differentiable mathematical formulation that accurately captures the entire drivable envelope. This novel model and formulation allow for the direct integration of dynamic feasibility and safety constraints into a unified planning and control framework, thereby removing the necessity for predefined references. The challenge of envelope planning, which refers to maximally approximating the safe drivable area, is tackled by combining reinforcement learning with optimization techniques. The framework is validated through both simulations and real world experiments, demonstrating its high performance across a variety of tasks, including racing, emergency collision avoidance and off road navigation. These results highlight the framework's scalability and broad applicability across a diverse set of scenarios.

This paper presents a novel terrain-adaptive local trajectory planner designed for the autonomous operation of off-road vehicles on deformable terrains. State-of-the-art solutions either do not account for deformable terrains, or do not offer sufficient robustness or computational speed. To bridge this research gap, the paper introduces a novel model predictive control (MPC) formulation. In contrast to the prevailing state-of-the-art approaches that rely exclusively on hard or soft constraints for obstacle avoidance, the present formulation enhances robustness by incorporating both types of constraints. The effectiveness and robustness of the formulation are evaluated through extensive simulations, encompassing a wide range of randomized scenarios, and compared against state-of-the-art methods. Subsequently, the formulation is augmented with an optimal-control-oriented terramechanics model from the literature, explicitly addressing terrain deformation. Additionally, a terrain estimator employing the unscented Kalman filter is utilized to dynamically adjust the sinkage exponent online, resulting in a terrain-adaptive formulation. This formulation is tested on a physical vehicle in real world experiments against a rigid-terrain formulation as the benchmark. The results showcase the superior safety and performance achieved by the proposed formulation, underscoring the critical significance of integrating terramechanics knowledge into the planning process. Specifically, the proposed terrain-adaptive formulation achieves reduced mean absolute sideslip angle, decreased mean absolute yaw rate, shorter time to goal, and a higher success rate, primarily attributed to its enhanced understanding of terramechanics within the planner.

A novel hierarchical global trajectory planner is presented to allow highly dynamical nonholonomic off-road autonomous vehicles to achieve high mobility on 3D terrains. On complex terrains with uneven topology, designing safe and feasible vehicle trajectories often demands an understanding of the vehicle's dynamical and nonholonomic constraints. Prior research, however, treats the global planning problem as a path planning problem without effectively accounting for topology or dynamical constraints. To address this gap, this paper presents a three-phase trajectory planning algorithm composed of an A*, a rapidly exploring random tree (RRT), and a local trajectory refining (LTR) phase to incorporate dynamical and nonholonomic constraints on uneven terrain. The algorithm is tested in scenarios with randomized terrain fields and obstacles to demonstrate the necessity for all three phases. The algorithm is shown to have lower cost, higher success rate, and higher computational efficiency compared to state-of-the-art methods. The algorithm is then tested by controlling a simulated MRZR vehicle on a 3D terrain along with a local controller, with comparisons to state-of-the-art algorithms. It is demonstrated that the new algorithm is capable of planning dynamically feasible trajectories with lower cost where the state-of-the-art algorithms fail to perform due to neglecting dynamical vehicle limitations.

Terrain traversability is critical for developing Go/No-Go maps for ground vehicles, which significantly impact a mission’s success. To predict the mobility of terrain, one must understand the soil characteristics. In-situ measurements performed in the field are the current method of collecting this information, which is time-consuming, costly, and can be lethal for military operations. This paper investigates an alternative approach using thermal, multispectral, and hyperspectral remote sensing from an unmanned aerial vehicle (UAV) platform. Remotely sensed data combined with machine learning (linear, ridge, lasso, partial least squares (PLS), support vector machines (SVM), and k nearest neighbors (KNN)) and deep learning (multi-layer perceptron (MLP) and convolutional neural network (CNN)) are used to perform a comparative study to estimate the soil properties, such as the soil moisture and terrain strength, used to generate prediction maps of these terrain characteristics. This study found that deep learning outperformed machine learning. Specifically, a multi-layer perceptron performed the best for predicting the percent moisture content (R2/RMSE = 0.97/1.55) and the soil strength (in PSI), as measured by a cone penetrometer for the averaged 0–6” (CP06) (R2/RMSE = 0.95/67) and 0–12” depth (CP12) (R2/RMSE = 0.92/94). A Polaris MRZR vehicle was used to test the application of these prediction maps for mobility purposes, and correlations were observed between the CP06 and the rear wheel slip and the CP12 and the vehicle speed. Thus, this study demonstrates the potential of a more rapid, cost-efficient, and safer approach to predict terrain properties for mobility mapping using remote sensing data with machine and deep learning algorithms.

Workload management is a critical concern in shared control of unmanned ground vehicles. In response to this challenge, prior studies have developed methods to estimate human operators’ workload by analyzing their physiological data. However, these studies have primarily adopted a single-model-single-feature or a single-model-multiple-feature approach. The present study proposes a Bayesian inference model to estimate workload, which leverages different machine learning models for different features. We conducted a human subject experiment with 24 participants, in which a human operator teleoperated a simulated High Mobility Multipurpose Wheeled Vehicle (HMMWV) with the help from an autonomy while performing a surveillance task simultaneously. Participants’ eye-related features, including gaze trajectory and pupil size change, were used as the physiological input to the proposed Bayesian inference model. Results show that the Bayesian inference model achieves a 0.823 F1 score, 0.824 precision, and 0.821 recall, outperforming the single models.

This paper is focused on the pulse-and-glide (PnG) control for a parallel hybrid electric vehicle (HEV) in car-following. PnG is an eco-driving technique that alternately turns on and off the engine to save fuel. For HEVs, the ride comfort concerns by PnG in conventional vehicles can be alleviated by oscillating the battery state of charge (SOC), instead of the vehicle speed. However, due to the ohmic losses during the battery charging and discharging process, the fuel saving potential is much reduced. In order to reach a balance between ride comfort and fuel economy, a control method is proposed to utilize both the vehicle body and the battery as the energy buffers simultaneously in the PnG operation. In particular, two minimum-time control problems, one for the pulsing phase and another for the gliding phase, are formulated and implemented as model predictive control (MPC). With the applications of a series of approximations and sparsity optimization techniques, these two control problems become quadratic programming (QP) problems after the application of the pseudo-spectral (PS) method. The online implementability is thus ensured. Numerical simulations using naturalistic driving data show on average 17.1% and 5.1% improvements of fuel economy for local and highway speeds, respectively. It is shown that the proposed method is able to improve the fuel economy while maintaining both ride comfort and SOC sustenance.