MetaSym: A Symplectic Meta-learning Framework for Physical Intelligence

Scalable and generalizable physics-aware deep learning has long been considered a significant challenge with various applications across diverse domains ranging from robotics to molecular dynamics. Central to almost all physical systems are symplectic forms, the geometric backbone that underpins fundamental invariants like energy and momentum. In this work, we introduce a novel deep learning architecture, MetaSym. In particular, MetaSym combines a strong symplectic inductive bias obtained from a symplectic encoder and an autoregressive decoder with meta-attention. This principled design ensures that core physical invariants remain intact while allowing flexible, data-efficient adaptation to system heterogeneities. We benchmark MetaSym on highly varied datasets such as a high-dimensional spring mesh system (Otness et al., 2021), an open quantum system with dissipation and measurement backaction, and robotics-inspired quadrotor dynamics. Our results demonstrate superior performance in modeling dynamics under few-shot adaptation, outperforming state-of-the-art baselines with far larger models.

Discovery of skill switching criteria for learning agile quadruped locomotion

This paper develops a hierarchical learning and optimization framework that can learn and achieve well-coordinated multi-skill locomotion. The learned multi-skill policy can switch between skills automatically and naturally in tracking arbitrarily positioned goals and recover from failures promptly. The proposed framework is composed of a deep reinforcement learning process and an optimization process. First, the contact pattern is incorporated into the reward terms for learning different types of gaits as separate policies without the need for any other references. Then, a higher level policy is learned to generate weights for individual policies to compose multi-skill locomotion in a goal-tracking task setting. Skills are automatically and naturally switched according to the distance to the goal. The proper distances for skill switching are incorporated in reward calculation for learning the high level policy and updated by an outer optimization loop as learning progresses. We first demonstrated successful multi-skill locomotion in comprehensive tasks on a simulated Unitree A1 quadruped robot. We also deployed the learned policy in the real world showcasing trotting, bounding, galloping, and their natural transitions as the goal position changes. Moreover, the learned policy can react to unexpected failures at any time, perform prompt recovery, and resume locomotion successfully. Compared to discrete switch between single skills which failed to transition to galloping in the real world, our proposed approach achieves all the learned agile skills, with smoother and more continuous skill transitions.

Offline Adaptation of Quadruped Locomotion using Diffusion Models

We present a diffusion-based approach to quadrupedal locomotion that simultaneously addresses the limitations of learning and interpolating between multiple skills and of (modes) offline adapting to new locomotion behaviours after training. This is the first framework to apply classifier-free guided diffusion to quadruped locomotion and demonstrate its efficacy by extracting goal-conditioned behaviour from an originally unlabelled dataset. We show that these capabilities are compatible with a multi-skill policy and can be applied with little modification and minimal compute overhead, i.e., running entirely on the robots onboard CPU. We verify the validity of our approach with hardware experiments on the ANYmal quadruped platform.

Constrained Skill Discovery: Quadruped Locomotion with Unsupervised Reinforcement Learning

Representation learning and unsupervised skill discovery can allow robots to acquire diverse and reusable behaviors without the need for task-specific rewards. In this work, we use unsupervised reinforcement learning to learn a latent representation by maximizing the mutual information between skills and states subject to a distance constraint. Our method improves upon prior constrained skill discovery methods by replacing the latent transition maximization with a norm-matching objective. This not only results in a much a richer state space coverage compared to baseline methods, but allows the robot to learn more stable and easily controllable locomotive behaviors. We successfully deploy the learned policy on a real ANYmal quadruped robot and demonstrate that the robot can accurately reach arbitrary points of the Cartesian state space in a zero-shot manner, using only an intrinsic skill discovery and standard regularization rewards.

InteLiPlan: Interactive Lightweight LLM-Based Planner for Domestic Robot Autonomy

We introduce a lightweight LLM-based framework designed to enhance the autonomy and robustness of domestic robots, targeting onboard embodied intelligence. By addressing challenges such as kinematic constraints and dynamic environments, our approach reduces reliance on large-scale data and incorporates a robot-agnostic pipeline. Our framework, InteLiPlan, ensures that the LLM model's decision-making capabilities are effectively aligned with robotic functions, enhancing operational robustness and adaptability, while our human-in-the-loop mechanism allows for real-time human intervention in the case where the system fails. We evaluate our method in both simulation and on the real Toyota HSR robot. The results show that our method achieves a 93% success rate in the fetch me task completion with system failure recovery, outperforming the baseline method in a domestic environment. InteLiPlan achieves comparable performance to the state-of-the-art large-scale LLM-based robotics planner, while guaranteeing real-time onboard computing with embodied intelligence.

Adaptive Manipulation using Behavior Trees

Many manipulation tasks use instances of a set of common motions, such as a twisting motion for tightening or loosening a valve. However, different instances of the same motion often require different environmental parameters (e.g. force/torque level), and thus different manipulation strategies to successfully complete; for example, grasping a valve handle from the side rather than head-on to increase applied torque. Humans can intuitively adapt their manipulation strategy to best suit such problems, but representing and implementing such behaviors for robots remains an open question. We present a behavior tree-based approach for adaptive manipulation, wherein the robot can reactively select from and switch between a discrete set of manipulation strategies during task execution. Furthermore, our approach allows the robot to learn from past attempts to optimize performance, for example learning the optimal strategy for different task instances. Our approach also allows the robot to preempt task failure and either change to a more feasible strategy or safely exit the task before catastrophic failure occurs. We propose a simple behavior tree design for general adaptive robot behavior and apply it in the context of industrial manipulation. The adaptive behavior outperformed all baseline behaviors that only used a single manipulation strategy, markedly reducing the number of attempts and overall time taken to complete the example tasks. Our results demonstrate potential for improved robustness and efficiency in task completion, reducing dependency on human supervision and intervention.

Gaitor: Learning a Unified Representation Across Gaits for Real-World Quadruped Locomotion

The current state-of-the-art in quadruped locomotion is able to produce robust motion for terrain traversal but requires the segmentation of a desired robot trajectory into a discrete set of locomotion skills such as trot and crawl. In contrast, in this work we demonstrate the feasibility of learning a single, unified representation for quadruped locomotion enabling continuous blending between gait types and characteristics. We present Gaitor, which learns a disentangled representation of locomotion skills, thereby sharing information common to all gait types seen during training. The structure emerging in the learnt representation is interpretable in that it is found to encode phase correlations between the different gait types. These can be leveraged to produce continuous gait transitions. In addition, foot swing characteristics are disentangled and directly addressable. Together with a rudimentary terrain encoding and a learned planner operating in this structured latent representation, Gaitor is able to take motion commands including desired gait type and characteristics from a user while reacting to uneven terrain. We evaluate Gaitor in both simulated and real-world settings on the ANYmal C platform. To the best of our knowledge, this is the first work learning such a unified and interpretable latent representation for multiple gaits, resulting in on-demand continuous blending between different locomotion modes on a real quadruped robot.

Curriculum-Based Reinforcement Learning for Quadrupedal Jumping: A Reference-free Design

Deep reinforcement learning (DRL) has emerged as a promising solution to mastering explosive and versatile quadrupedal jumping skills. However, current DRL-based frameworks usually rely on well-defined reference trajectories, which are obtained by capturing animal motions or transferring experience from existing controllers. This work explores the possibility of learning dynamic jumping without imitating a reference trajectory. To this end, we incorporate a curriculum design into DRL so as to accomplish challenging tasks progressively. Starting from a vertical in-place jump, we then generalize the learned policy to forward and diagonal jumps and, finally, learn to jump across obstacles. Conditioned on the desired landing location, orientation, and obstacle dimensions, the proposed approach contributes to a wide range of jumping motions, including omnidirectional jumping and robust jumping, alleviating the effort to extract references in advance. Particularly, without constraints from the reference motion, a 90cm forward jump is achieved, exceeding previous records for similar robots reported in the existing literature. Additionally, continuous jumping on the soft grassy floor is accomplished, even when it is not encountered in the training stage. A supplementary video showing our results can be found at https://youtu.be/nRaMCrwU5X8 .

Towards Agility: A Momentum Aware Trajectory Optimisation Framework using Full-Centroidal Dynamics & Implicit Inverse Kinematics

Online planning and execution of acrobatic maneuvers pose significant challenges in legged locomotion. Their underlying combinatorial nature, along with the current hardware's limitations constitute the main obstacles in unlocking the true potential of legged-robots. This letter tries to expose the intricacies of these optimal control problems in a tangible way, directly applicable to the creation of more efficient online trajectory optimisation frameworks. By analysing the fundamental principles that shape the behaviour of the system, the dynamics themselves can be exploited to surpass its hardware limitations. More specifically, a trajectory optimisation formulation is proposed that exploits the system's high-order nonlinearities, such as the nonholonomy of the angular momentum, and phase-space symmetries in order to produce feasible high-acceleration maneuvers. By leveraging the full-centroidal dynamics of the quadruped ANYmal C and directly optimising its footholds and contact forces, the framework is capable of producing efficient motion plans with low computational overhead. The feasibility of the produced trajectories is ensured by taking into account the configuration-dependent inertial properties of the robot during the planning process, while its robustness is increased by supplying the full analytic derivatives & hessians to the solver. Finally, a significant portion of the discussion is centred around the deployment of the proposed framework on the ANYmal C platform, while its true capabilities are demonstrated through real-world experiments, with the successful execution of high-acceleration motion scenarios like the squat-jump.

R-LGP: A Reachability-guided Logic-geometric Programming Framework for Optimal Task and Motion Planning on Mobile Manipulators

This paper presents an optimization-based solution to task and motion planning (TAMP) on mobile manipulators. Logic-geometric programming (LGP) has shown promising capabilities for optimally dealing with hybrid TAMP problems that involve abstract and geometric constraints. However, LGP does not scale well to high-dimensional systems (e.g. mobile manipulators) and can suffer from obstacle avoidance issues. In this work, we extend LGP with a sampling-based reachability graph to enable solving optimal TAMP on high-DoF mobile manipulators. The proposed reachability graph can incorporate environmental information (obstacles) to provide the planner with sufficient geometric constraints. This reachability-aware heuristic efficiently prunes infeasible sequences of actions in the continuous domain, hence, it reduces replanning by securing feasibility at the final full trajectory optimization. Our framework proves to be time-efficient in computing optimal and collision-free solutions, while outperforming the current state of the art on metrics of success rate, planning time, path length and number of steps. We validate our framework on the physical Toyota HSR robot and report comparisons on a series of mobile manipulation tasks of increasing difficulty.