Abstract-- This paper investigates a sample-based solution to the hybrid mode control problem across non-differentiable and algorithmic hybrid modes. Our approach reasons about a set of hybrid control modes as an integer-based optimization problem where we select what mode to apply, when to switch to another mode, and the duration for which we are in a given control mode. A sample-based variation is derived to efficiently search the integer domain for optimal solutions. We find our formulation yields strong performance guarantees that can be applied to a number of robotics-related tasks. In addition, our approach is able to synthesize complex algorithms and policies to compound behaviors and achieve challenging tasks. Last, we demonstrate the effectiveness of our approach in a real-world robotic examples that requires reactive switching between long-term planning and high-frequency control. I. INTRODUCTION Modern agile robotic systems must dynamically switch between discrete modes--such as making and breaking contacts--to synthesize complex behaviors like locomotion and manipulation.

Imitation learning has driven the development of generalist policies capable of autonomously solving multiple tasks. However, when a pretrained policy makes errors during deployment, there are limited mechanisms for users to correct its behavior. While collecting additional data for finetuning can address such issues, doing so for each downstream use case is inefficient at deployment. My research proposes an alternative: keeping pretrained policies frozen as a fixed skill repertoire while allowing user interactions to guide behavior generation toward user preferences at inference time. By making pretrained policies steerable, users can help correct policy errors when the model struggles to generalize-without needing to finetune the policy. Specifically, I propose (1) inference-time steering, which leverages user interactions to switch between discrete skills, and (2) task and motion imitation, which enables user interactions to edit continuous motions while satisfying task constraints defined by discrete symbolic plans. These frameworks correct misaligned policy predictions without requiring additional training, maximizing the utility of pretrained models while achieving inference-time user objectives.

Bb-Simplex is centered around the Simplex control architecture, which consists of a high-performance advanced controller that is not guaranteed to maintain safety of the plant, a verified-safe baseline controller, and a decision module that switches control of the plant between the two controllers to ensure safety without sacrificing performance. In Bb-Simplex, Barrier certificates are used to prove that the baseline controller ensures safety. Furthermore, Bb-Simplex features a new automated method for deriving, from the barrier certificate, the conditions for switching between the controllers. Our method is based on the Taylor expansion of the barrier certificate and yields computationally inexpensive switching conditions. We also propose extensions to Bb-Simplex to enable its use in hybrid systems, which have multiple modes each with its own dynamics, and to support its use when only approximate dynamics (not exact dynamics) are available, for both continuous-time and hybrid dynamical systems. We consider significant applications of Bb-Simplex to microgrids featuring advanced controllers in the form of neural networks trained using reinforcement learning. These microgrids are modeled in RTDS, an industry-standard high-fidelity, real-time power systems simulator. Our results demonstrate that Bb-Simplex can automatically derive switching conditions for complex continuous-time and hybrid systems, the switching conditions are not overly conservative, and Bb-Simplex ensures safety even in the presence of adversarial attacks on the neural controller when only approximate dynamics (with an error bound) are available.

Loco-manipulation calls for effective whole-body control and contact-rich interactions with the object and the environment. Existing learning-based control frameworks rely on task-specific engineered rewards, training a set of low-level skill policies and explicitly switching between them with a high-level policy or FSM, leading to quasi-static and fragile transitions between skills. In contrast, for solving highly dynamic tasks such as soccer, the robot should run towards the ball, decelerating into an optimal approach configuration to seamlessly switch to dribbling and eventually score a goal - a continuum of smooth motion. To this end, we propose to learn a single Oracle Guided Multi-mode Policy (OGMP) for mastering all the required modes and transition maneuvers to solve uni-object bipedal loco-manipulation tasks. Specifically, we design a multi-mode oracle as a closed loop state-reference generator, viewing it as a hybrid automaton with continuous reference generating dynamics and discrete mode jumps. Given such an oracle, we then train an OGMP through bounded exploration around the generated reference. Furthermore, to enforce the policy to learn the desired sequence of mode transitions, we present a novel task-agnostic mode-switching preference reward that enhances performance. The proposed approach results in successful dynamic loco-manipulation in omnidirectional soccer and box-moving tasks with a 16-DoF bipedal robot HECTOR. Supplementary video results are available at https://www.youtube.com/watch?v=gfDaRqobheg

Grounding the common-sense reasoning of Large Language Models (LLMs) in physical domains remains a pivotal yet unsolved problem for embodied AI. Whereas prior works have focused on leveraging LLMs directly for planning in symbolic spaces, this work uses LLMs to guide the search of task structures and constraints implicit in multi-step demonstrations. Specifically, we borrow from manipulation planning literature the concept of mode families, which group robot configurations by specific motion constraints, to serve as an abstraction layer between the high-level language representations of an LLM and the low-level physical trajectories of a robot. By replaying a few human demonstrations with synthetic perturbations, we generate coverage over the demonstrations' state space with additional successful executions as well as counterfactuals that fail the task. Our explanation-based learning framework trains an end-to-end differentiable neural network to predict successful trajectories from failures and as a by-product learns classifiers that ground low-level states and images in mode families without dense labeling. The learned grounding classifiers can further be used to translate language plans into reactive policies in the physical domain in an interpretable manner. We show our approach improves the interpretability and reactivity of imitation learning through 2D navigation and simulated and real robot manipulation tasks. Website: https://yanweiw.github.io/glide

This study focuses on a layered, experience-based, multi-modal contact planning framework for agile quadrupedal locomotion over a constrained rebar environment. To this end, our hierarchical planner incorporates locomotion-specific modules into the high-level contact sequence planner and solves kinodynamically-aware trajectory optimization as the low-level motion planner. Through quantitative analysis of the experience accumulation process and experimental validation of the kinodynamic feasibility of the generated locomotion trajectories, we demonstrate that the experience planning heuristic offers an effective way of providing candidate footholds for a legged contact planner. Additionally, we introduce a guiding torso path heuristic at the global planning level to enhance the navigation success rate in the presence of environmental obstacles. Our results indicate that the torso-path guided experience accumulation requires significantly fewer offline trials to successfully reach the goal compared to regular experience accumulation. Finally, our planning framework is validated in both dynamics simulations and real hardware implementations on a quadrupedal robot provided by Skymul Inc.

In prior work, learning from demonstration (LfD) [1, 2] has successfully enabled robots to accomplish multi-step tasks by segmenting demonstrations (primarily of robot end-effector or tool trajectories) into sub-tasks/goals [3, 4, 5, 6, 7, 8], phases [9, 10], keyframes [11, 12], or skills/primitives/options [13, 14, 15, 16]. Most of these abstractions assume reaching subgoals sequentially will deliver the desired outcomes; however, successful imitation of many manipulation tasks with spatial/temporal constraints cannot be reduced to imitation at the motion level unless the learned motion policy also satisfies these constraints. This becomes highly relevant if we want robots to not only imitate but also generalize, adapt and be robust to perturbations imposed by humans, who are in the loop of task learning and execution. LfD techniques that learn stable motion policies with convergence guarantees (e.g., Dynamic Movement Primitives (DMP) [17], Dynamical Systems (DS) [18]) are capable of providing such desired properties but only at the motion level. As shown in Figure 1 (a-b) a robot can successfully replay a soup-scooping task while being robust to physical perturbations with a learned DS. Nevertheless, if the spoon orientation is perturbed to a state where all material is dropped, as seen in Figure 1 (c), the motion policy will still lead the robot to the target, unaware of the task-level failure or how to recover from it.

The rise of AI in human contexts places new demands on automated systems to be transparent and explainable. We examine some anthropomorphic ideas and principles relevant to such accountablity in order to develop a theoretical framework for thinking about digital systems in complex human contexts and the problem of explaining their behaviour. Structurally, systems are made of modular and hierachical components, which we abstract in a new system model using notions of modes and mode transitions. A mode is an independent component of the system with its own objectives, monitoring data, and algorithms. The behaviour of a mode, including its transitions to other modes, is determined by functions that interpret each mode's monitoring data in the light of its objectives and algorithms. We show how these belief functions can help explain system behaviour by visualising their evaluation as trajectories in higher-dimensional geometric spaces. These ideas are formalised mathematically by abstract and concrete simplicial complexes. We offer three techniques - a framework for design heuristics, a general system theory based on modes, and a geometric visualisation - and apply them in three types of human-centred systems.

In industrial systems, certain process variables that need to be monitored for detecting faults are often difficult or impossible to measure. Soft sensor techniques are widely used to estimate such difficult-to-measure process variables from easy-to-measure ones. Soft sensor modeling requires training datasets including the information of various states such as operation modes, but the fault dataset with the target variable is insufficient as the training dataset. This paper describes a semi-supervised approach to soft sensor modeling to incorporate an incomplete dataset without the target variable in the training dataset. To incorporate the incomplete dataset, we consider the properties of processes at transition points between operation modes in the system. The regression coefficients of the operation modes are estimated under constraint conditions obtained from the information on the mode transitions. In a case study, this constrained soft sensor modeling was used to predict refrigerant leaks in air-conditioning systems with heating and cooling operation modes. The results show that this modeling method is promising for soft sensors in a system with multiple operation modes.

Innovative methods have been developed for diagnosis, activity monitoring, and state estimation that achieve high accuracy through the use of stochastic models involving hybrid discrete and continuous behaviors. A key bottleneck is the automated acquisition of these hybrid models, and recent methods have focused predominantly on Jump Markov processes and piecewise autoregressive models. In this paper, we present a novel algorithm capable of performing unsupervised learning of guarded Probabilistic Hybrid Automata (PHA) models, which extends prior work by allowing stochastic discrete mode transitions in a hybrid system to have a functional dependence on its continuous state. Our experiments indicate that guarded PHA models can yield significant performance improvements when used by hybrid state estimators, particularly when diagnosing the true discrete mode of the system, without any noticeable impact on their real-time performance.