Recent studies on Vision-Language-Action (VLA) models have shifted from the end-to-end action-generation paradigm toward a pipeline involving task planning followed by action generation, demonstrating improved performance on various complex, long-horizon manipulation tasks. However, existing approaches vary significantly in terms of network architectures, planning paradigms, representations, and training data sources, making it challenging for researchers to identify the precise sources of performance gains and determine which component is more difficult to learn. To systematically investigate the impacts of different planning paradigms and representations isolating from network architectures and training data, in this paper, we introduce \name, a unified VLA architecture suite capable of various task planning paradigms, and design a comprehensive suite of controlled experiments across diverse object categories (rigid and deformable), visual modalities (2D and 3D), environments (simulation and real-world), and end-effectors (grippers and dexterous hands). Our results demonstrate that: 1) visually grounded planning representations are generally better than language planning representations; 2) the Hierarchical-VLA paradigm generally achieves superior performance than other paradigms, albeit at the cost of slower training and inference speeds.

Plasticity, the ability of a neural network to evolve with new data, is crucial for high-performance and sample-efficient visual reinforcement learning (VRL). Although methods like resetting and regularization can potentially mitigate plasticity loss, the influences of various components within the VRL framework on the agent's plasticity are still poorly understood. In this work, we conduct a systematic empirical exploration focusing on three primary underexplored facets and derive the following insightful conclusions: (1) data augmentation is essential in maintaining plasticity; (2) the critic's plasticity loss serves as the principal bottleneck impeding efficient training; and (3) without timely intervention to recover critic's plasticity in the early stages, its loss becomes catastrophic. These insights suggest a novel strategy to address the high replay ratio (RR) dilemma, where exacerbated plasticity loss hinders the potential improvements of sample efficiency brought by increased reuse frequency. Rather than setting a static RR for the entire training process, we propose Adaptive RR, which dynamically adjusts the RR based on the critic’s plasticity level. Extensive evaluations indicate that Adaptive RR not only avoids catastrophic plasticity loss in the early stages but also benefits from more frequent reuse in later phases, resulting in superior sample efficiency.

To substantially enhance robot intelligence, there is a pressing need to develop a large model that enables general-purpose robots to proficiently undertake a broad spectrum of manipulation tasks, akin to the versatile task-planning ability exhibited by LLMs. The vast diversity in objects, robots, and manipulation tasks presents huge challenges. Our work introduces a comprehensive framework to develop a foundation model for general robotic manipulation that formalizes a manipulation task as contact synthesis. Specifically, our model takes as input object and robot manipulator point clouds, object physical attributes, target motions, and manipulation region masks. It outputs contact points on the object and associated contact forces or post-contact motions for robots to achieve the desired manipulation task. We perform extensive experiments both in the simulation and real-world settings, manipulating articulated rigid objects, rigid objects, and deformable objects that vary in dimensionality, ranging from one-dimensional objects like ropes to two-dimensional objects like cloth and extending to three-dimensional objects such as plasticine. Our model achieves average success rates of around 90%. Supplementary materials and videos are available on our project website at https://manifoundationmodel.github.io/.

Based on actor-critic neural networks (NNs), an optimal controller is proposed for solving the constrained control problem of an affine nonlinear discrete-time system with disturbances. The actor NNs provide the control signals and the critic NNs work as the performance indicators of the controller. By converting the original state constraints into new input constraints and state constraints, the penalty functions are introduced into the cost function, and then the constrained optimal control problem is transformed into an unconstrained one. Further, the relationship between the optimal control input and worst-case disturbance is obtained using the Game theory. With Lyapunov stability theory, the control signals are ensured to be uniformly ultimately bounded (UUB). Finally, the effectiveness of the control algorithms is tested through a numeral simulation using a third-order dynamic system.