This work presents a novel Learning Model Predictive Control (LMPC) strategy for autonomous racing at the handling limit that can iteratively explore and learn unknown dynamics in high-speed operational domains. We start from existing LMPC formulations and modify the system dynamics learning method. In particular, our approach uses a nominal, global, nonlinear, physics-based model with a local, linear, data-driven learning of the error dynamics. We conducted experiments in simulation and on 1/10th scale hardware, and deployed the proposed LMPC on a full-scale autonomous race car used in the Indy Autonomous Challenge (IAC) with closed loop experiments at the Putnam Park Road Course in Indiana, USA. The results show that the proposed control policy exhibits improved robustness to parameter tuning and data scarcity. Incremental and safety-aware exploration toward the limit of handling and iterative learning of the vehicle dynamics in high-speed domains is observed both in simulations and experiments.

In this work, we propose a numerical method for the solution of local generalized Nash equilibria (GNE) for the class of open-loop general-sum dynamic games for agents with nonlinear dynamics and constraints. In particular, we formulate a sequential quadratic programming (SQP) approach which requires only the solution of a single convex quadratic program at each iteration and is locally convergent. Central to the effectiveness of our approach is a non-monotonic line search method and a novel merit function for SQP step acceptance which helps to improve solver convergence beyond the local neighborhood of a GNE. We demonstrate the effectiveness of the algorithm in the context of car racing, where we see up to 32% improvement of success rate when comparing against a recent solution approach for dynamic games. We also make our code available at https://github.com/zhu-edward/DGSQP.

In head-to-head racing, performing tightly con-strained, but highly rewarding maneuvers, such as overtaking, require an accurate model of interactive behavior of the opposing target vehicle (TV). We propose to construct a prediction model given data of the TV from previous races. In particular, a one-step Gaussian process (GP) model is trained on closed-loop interaction data to learn the behavior of a TV driven by an unknown policy. Predictions of the nominal trajectory and associated uncertainty are rolled out via a sampling-based approach and are used in a model predictive control (MPC) policy for the ego vehicle in order to intelligently trade-off between safety and performance when racing against a TV. In a Monte Carlo study, we compare the GP-based predictor in closed-loop with the MPC policy against several predictors from literature and observe that the GP-based predictor achieves similar win rates while maintaining safety in up to 3x more races. Through experiments, we demonstrate the approach in real-time on a 1/10th scale racecar platform operating at speeds of around 2.8 m/s, and show a significant level of improvement when using the GP-based predictor over a baseline MPC predictor. Videos of the experiments can be found at https://voutu.be/KMSs4ofDfIs.

We present a hierarchical control approach for maneuvering an autonomous vehicle (AV) in tightly-constrained environments where other moving AVs and/or human driven vehicles are present. A two-level hierarchy is proposed: a high-level data-driven strategy predictor and a lower-level model-based feedback controller. The strategy predictor maps an encoding of a dynamic environment to a set of high-level strategies via a neural network. Depending on the selected strategy, a set of time-varying hyperplanes in the AV’s position space is generated online and the corresponding halfspace constraints are included in a lower-level model-based receding horizon controller. These strategy-dependent constraints drive the vehicle towards areas where it is likely to remain feasible. Moreover, the predicted strategy also informs switching between a discrete set of policies, which allows for more conservative behavior when prediction confidence is low. We demonstrate the effectiveness of the proposed data-driven hierarchical control framework in a two-car collision avoidance scenario through simulations and experiments on a 1/10 scale autonomous car platform where the strategy-guided approach outperforms a model predictive control baseline in both cases.

This paper presents a new teleoperated spherical tensegrity robot capable of performing locomotion on steep inclined surfaces. With a novel control scheme centered around the simultaneous actuation of multiple cables, the robot demonstrates robust climbing on inclined surfaces in hardware experiments and speeds significantly faster than previous spherical tensegrity models. This robot is an improvement over other iterations in the TT-series and the first tensegrity to achieve reliable locomotion on inclined surfaces of up to 24°. We analyze locomotion in simulation and hardware under single and multi-cable actuation, and introduce two novel multi-cable actuation policies, suited for steep incline climbing and speed, respectively. We propose compelling justifications for the increased dynamic ability of the robot and motivate development of optimization algorithms able to take advantage of the robot's increased control authority.

This work presents a 10 kg tensegrity ball probe that can quickly and precisely deliver a 1 kg payload over a 1 km distance on the Moon by combining cable-driven rolling and thruster-based hopping. Previous research has shown that cable-driven rolling is effective for precise positioning, even in rough terrain. However, traveling large distances using thruster-based hopping, which is made feasible by the lightweight and compliant nature of the tensegrity structure, has not been explored. To evaluate the feasibility of a thruster-based tensegrity robot, a centrally-positioned cold gas thruster with nitrogen propellant was selected, and the system was simulated using the NASA Tensegrity Robotics Toolkit (NTRT) for four hopping profiles on hilly terrains. Optimizing energy efficiency and mechanical capabilities of the tensegrity robot, hopping profiles with a long flight distance per hop, followed by the higher accuracy rolling, are recommended. Simulations also show that thrust regulation can improve energy efficiency. Regulation of thrust magnitude can be achieved using a pressure regulator, but regulation of thrust orientation calls for additional control effort. In this paper, it is demonstrated that gimbal systems as well as shape-shifting control of the tensegrity structure have the potential to regulate thrust orientation. Finally, algorithms for localization and path planning that combine hopping and rolling for energy-efficient navigation are presented.