Searcharxiv⌕ Search

arXiv subjects

Caicheng Wang

Publications and source records attributed to Caicheng Wang.

5 recordsLinked to original sources

Meituan-Robotics-0: A Vision-Language-Action Foundation Model for Desktop Manipulation

We present Meituan-Robotics-0 (MR0), a vision-language-action foundation model and embodied-agent system for stationary-base desktop manipulation. MR0 combines 147 million vision-language samples with 15,060 hours of curated robot trajectories spanning 12 embodiments. A three-stage training recipe aligns multimodal representations with robot behavior through discrete action-token pre-training, transfers these priors to continuous cross-embodiment control through flow-matching mid-training, and adapts the policy through real-world post-training. We introduce Rollback, which pairs replaced policy segments with human corrections to provide positive and negative supervision for CFG-DAgger without value learning or advantage labeling. A unified end-effector interface supports heterogeneous embodiments, while asynchronous action-chunk merging maintains smooth execution under variable inference latency. An event-driven agent maintains persistent task events and reassesses plans when progress or feasibility changes, enabling cross-task retrieval and conditional task resumption. MR0 achieves 93.6% success on RoboTwin 2.0 and ranks fifth in the RoboDojo VLA track. Under matched HG-DAgger post-training, MR0 achieves higher average success rate and progress score than $π_{0.5}$ across three real-robot tasks. On Ethernet insertion, CFG-DAgger trained with Rollback data reaches 70% success and a progress score of 0.90. Qualitative agent deployments demonstrate cross-task event retrieval and recovery from temporarily unsatisfied task conditions.

cs.RO↗

ClothDojo: Learned Data Generation and Benchmarking for Bimanual Garment Flattening and Folding

Garment flattening and folding require robots to identify garment structure during manipulation and adapt their actions as the cloth deforms. We present ClothDojo, a simulation platform that generates adaptive manipulation trajectories at scale across diverse garments and initial states and evaluates policies in closed loop. ClothDojo encodes each garment's canonical mesh coordinates as an NOCS texture, providing canonical surface-position cues through the visual interface of a pretrained vision-language-action model. We fine-tune the model on human demonstrations and corrections to obtain a data-generation policy. The policy autonomously generates simulated manipulation trajectories, and we retain rollouts that satisfy task-specific success criteria. Each retained trajectory is re-rendered under a randomly sampled appearance to provide RGB observations for downstream policies that do not receive NOCS. Across both tasks, the policy produces nearly 8,000 successful trajectories over hundreds of garments. RGB policies trained on these generated trajectories outperform those trained directly on the original human demonstrations and corrections across all evaluated garment splits and appearance settings. Evaluations across multiple policy models show that overall performance and appearance robustness vary across models and tasks, while generalization to unseen garments remains a common challenge.

cs.RO↗

SimPT: An Automated Simulation Framework for Intervention-Based VLA Post-Training

Human-in-the-loop (HIL) post-training improves vision-language-action (VLA) policies through expert corrections, but costly human supervision and repeated real-robot evaluation limit algorithm exploration and development. We introduce SimPT (Simulation-based Post-Training), a simulation framework that automates intervention-based VLA post-training by integrating policy rollouts, corrective data collection, policy training, and standalone evaluation. To automatically determine intervention timing and recovery strategies, we develop an Agent-Compiled Progress Graph that tracks task progress using privileged simulator state and retains stable snapshots during policy execution. After a rollout fails, a CuRobo-based expert attempts recovery from these snapshots, starting at more advanced task stages and falling back to earlier ones as needed to generate corrective trajectories. We implement and compare four post-training methods, RECAP, HG-DAgger, DPO+SFT, and FlowPRO, across seven RoboTwin manipulation tasks. Relative to a fixed-interval snapshot baseline, the Progress Graph reduces recovery attempts by 75.0% and increases recovery success from 93.4% to 98.9%. Using SimPT, we develop RECAP-CF-LPF, a refinement of RECAP that combines cross-fitting and low-pass filtering of value predictions. This refinement increases the overall success rate from 76.50% to 79.36% in held-out simulation trials. In real-robot HIL post-training, it increases the success rate from 72% to 96% across 25 scenarios of an Ethernet cable-insertion task. These results demonstrate the utility of SimPT for exploring and developing VLA post-training algorithms.

cs.RO↗

A self-learning framework for adaptive free-form pipe routing in aeroengines

Pipe routing is a highly complex, time-consuming, and no-deterministic polynomial-time hard (NP-hard) problem in aeroengine design. Despite extensive research efforts in optimizing constant-curvature pipe routing, the growing demand for free-form pipes poses new challenges. Dynamic design environments and fuzzy layout rules further impact the optimization performance and efficiency. To tackle these challenges, this study proposes a self-learning-based method (SLPR) for optimizing free-form pipe routing in aeroengines. The SLPR is based on the proximal policy optimization (PPO) algorithm and integrates a unified rule modeling framework for efficient obstacle detection and fuzzy rule modeling in continuous space. Additionally, a potential energy table is constructed to enable rapid queries of layout tendencies and interference. The agent within SLPR iteratively refines pipe routing and accumulates the design knowledge through interaction with the environment. Once the design environment shifts, the agent can swiftly adapt by fine-tuning network parameters. Comparative tests reveal that SLPR ensures smooth pipe routing through cubic non-uniform B-spline (NURBS) curves, avoiding redundant pipe segments found in constant-curvature pipe routing. Results in both static and dynamic design environments demonstrate that SLPR outperforms three representative baselines in terms of the pipe length reduction, the adherence to layout rules, the path complexity, and the computational efficiency. Furthermore, tests in dynamic environments indicate that SLPR eliminates labor-intensive searches from scratch and even yields superior solutions compared to the retrained model. These results highlight the practical value of SLPR for real-world pipe routing, meeting lightweight, precision, and sustainability requirements of the modern aeroengine design.

cs.LG↗

Design for Manufacturing: A Manufacturability Knowledge-Integrated Reinforcement Learning Framework for Free-Form Pipe Routing in Aeroengines

Design for manufacturing plays a critical role in advanced aeroengine development, where complex components necessitate careful consideration of manufacturability. However, current practices in pipe routing remain largely decoupled from down-stream manufacturing, leading to labor-intensive, trial-and-error iterations to achieve manufacturable designs. To address this problem, this study proposes the Frenet-based pipe routing optimization (FPRO) framework, a manufacturability knowledge-integrated reinforcement learning approach for free-form pipe design in aeroengines. FPRO formulates the routing problem as a boundary value problem in the Frenet frame. In this framework, the pipe path is represented by curvature and torsion profiles, which are generated using cubic Hermite interpolation. To integrate design and manufacturing, domain-specific manufacturing knowledge is embedded as constraints on the permissible ranges of curvature and torsion. The path optimization is performed using the proximal policy optimization algorithm with stochastic exploration and a stage-guided reward mechanism. A unified mapping formulation then translates the optimized path into motion trajectories for the bending die, enabling direct fabrication on a six-axis free-bending machine. Experimental results demonstrate that FPRO consistently generates collision-free, manufacturable paths with smoother geometric profiles compared to Cartesian-based methods. It also achieves faster convergence and superior performance in terminal alignment, path length, obstacle avoidance, and manufacturability compared to state-of-the-art reinforcement learning baselines. Real-world validation confirms the close geometric correspondence between the manufactured pipe and its digital design, validating the practical feasibility of FPRO.

cs.LG↗