Technical Directions
Robotics coverage relating to manipulation.
471 selected reportsLast 30 days 407 recordsAll coverage 1,228 records
This study introduces PSA-PPO, a target-conditioned local reaching framework for robotic harvesting of hydroponic lettuce. Trained entirely in simulation, PSA-PPO achieves 94.7% reaching rate in physical testing, outperforming classical planners. The method integrates safety constraints and trajectory-based encoding to improve performance in constrained environments.
This release introduces the lerobot/so101-tabletop-mujoco environment within the LeRobot EnvHub framework. It simulates an SO-101 arm interacting with four colored cubes and a green box on a table using MuJoCo physics. The setup includes front and wrist cameras, 6 joint positions for observation, and absolute joint targets for control at 30 fps. A corresponding Isaac Sim version is also available, allowing the same policies to run across both simulators. The environment supports seeded resets, noise injection, and headless execution via EGL, all under an Apache-2.0 license.
The LeRobot team released lerobot/so101-tabletop-isaac, an environment built natively in NVIDIA Isaac Sim 6.2 featuring an SO-101 arm, four wooden cubes and a green box on a table. The layout, cameras, observations and actions match the MuJoCo version, allowing the same policies to run against both simulators. The environment is Apache-2.0 licensed, includes bundled assets requiring no network access at runtime, and runs on an NVIDIA RTX GPU with PhysX physics and RTX rendering.
This research presents PACT, a method that jointly optimizes the mechanical design and posture retargeting of a 17-servo robot hand to better replicate human grasping postures. It also achieves an 86% grasping success rate on YCB objects, with an imitation-learning policy achieving 75% success in a pick-and-place task.
This paper presents I3L, a system designed to address reliability gaps in offline-trained policies for long-horizon whole-body mobile manipulation. By combining whole-body correction with iterative policy refinement via a low-cost, open-source JoyLo+ interface, the authors demonstrate significant improvements in autonomous success rates across multiple tasks. Controlled simulation experiments show that four-round HG-DAgger achieves the highest final success, improving autonomous success by an average of 68.3 percentage points over base policies and achieving at least 90% success on every task. The system also reduces takeover duration per episode by 72% over four rounds, offering practical insights for collecting and using corrections in long-horizon tasks.
Researchers from the Shenyang Institute of Automation have developed a soft robotic fish that can swim at high speeds and perform agile three-dimensional maneuvers. The design integrates a rigid–flexible coupled propulsion system with dynamic buoyancy regulation, inspired by biological mechanisms such as muscle-skeleton coordination and swim bladder function. The robotic fish achieves a maximum speed of 2.13 body lengths per second and can perform complex 3D movements like turning, ascending, and spiral ascent, offering a potential approach for high-performance underwater robots.
MagicAtom's EchoPolicy-0 framework achieved a 75.4% average success rate in MolmoSpaces Combined benchmark, ranking first globally, while MagicBrain achieved 72.4% with a 94.4% grasp success rate. Both frameworks address challenges in VLA by integrating visual language models with geometric tools and action policies, improving task execution and generalization in complex environments.
This work introduces DexAgent, an agentic framework that transforms a single egocentric human video and task description into diverse robot training data. By reconstructing underlying physical interactions in simulation, verifying object properties, and generating optimized robot trajectories, DexAgent enables robots to learn complex manipulation tasks from one demonstration. Across eleven real-world tasks, policies trained on DexAgent-generated data achieve 63.6% average success compared with 18.2% for the strongest baseline. The framework organizes video-to-robot conversion into four stages: semantic understanding, object property-based simulation reconstruction, robot trajectory optimization, and robot data generation, supported by a shared tool library that grows across demonstrations.
FineART is a bimanual manipulation dataset co-released by Scale AI and Hugging Face LeRobot. The data was collected and annotated by the Physical AI unit at Scale AI. FineART is a bimanual manipulation dataset where every episode comes with dense, human-written subtask annotations. It has 40,543 teleoperated episodes (1,718 hours), 533,913 subtask labels, and 151 tasks, all collected on a Trossen Stationary AI (ALOHA-style) bimanual robot.
This paper presents PriMoRetarget, a framework for transferring human motion data to humanoid robots that addresses challenges in multi-character and object interactions. By replacing relative objective weighting with an explicit priority hierarchy, the method ensures that critical characteristics such as foot placement and interaction contacts take precedence over lower-priority pose objectives. Evaluated on four datasets covering single-character motion, character-object manipulation, and character-character interactions, the resulting motions better preserve task-relevant interactions and provide reference trajectories that can be tracked more reliably than those produced by widely used retargeting methods.
The Yuance Future team demonstrated their robot's performance in complex tasks, including reaching high places to retrieve objects, picking up bottles from low positions, clearing tables, and operating dishwashers. The article introduced their 'Whole-Body Intelligence' (Whole-Body Intelligence, WBI) technology approach, emphasizing the integration of body coordination and task execution. By collecting human motion data, real-machine training, and cross-body validation, the robot gradually adapts to various challenges in real-world scenarios.
AthenaZero, a new dual-arm robotic system developed by researchers at the Robotics and AI Institute, demonstrates agile, human-like movements in tasks such as throwing, catching, and batting. The robot's design incorporates low inertia and backdrivability, enabling more natural and controlled interactions with objects. The study explores the challenges in robotic manipulation and presents a novel approach to improving dexterous interaction with the environment.
This research introduces UniCross, a unified framework for dexterous manipulation that integrates four canonical skills—grasping, relocation, in-hand rotation, and in-hand translation—into a single relational task formulation. By modeling these skills with shared constraints and objectives, the approach achieves cross-skill compatibility, enabling distillation of a single policy without skill identifiers. The method supports long-horizon skill chaining and generalizes to unseen objects and hand morphologies, demonstrating that distinct manipulation skills can be instantiated from a common relational formulation.
This paper introduces MagCilia, a compact tactile sensor that integrates a flexible magnetic-cilia structure with a Hall sensor to enable three-dimensional force sensing. Using quasi-static finite element analysis and causal history fusion regression (CHFR), the system reconstructs forces from coupled magnetic channels with root-mean-square errors of 0.40 N, 0.57 N, and 0.69 N for Fx, Fy, and Fz respectively, achieving coefficients of determination between 0.90 and 0.93. Robotic experiments demonstrate the sensor's ability to guide gripper adjustments based on tangential forces and monitor multi-axis loads under external perturbations. Additionally, frequency-domain features allow the system to distinguish six surface categories with 99.39% accuracy in cross-validation grouped by acquisition session, and an online demonstration successfully identifies all six tested surfaces.
Researchers from ETH Zurich have developed a self-contained robotic hand that can walk on its fingertips and manipulate objects, using reinforcement learning and off-the-shelf hardware. The hand can navigate different surfaces, right itself, and push objects autonomously, suggesting new possibilities for modular and autonomous robotic systems.
TrinaTracker, a subsidiary of Trinasolar, has launched two robotic products—Buildex PV Module Installation Robot and Aurora PV Cleaning Robot—to enhance its Smart PV Ecosystem. These robots aim to automate and optimize solar farm construction and maintenance, reducing labor costs, improving precision, and increasing efficiency through AI and robotics integration.
In 2026, humanoid robots are transitioning from technology demonstrations to actual procurement in industries such as injection molding, 3C manufacturing, automotive, and healthcare. The article discusses the evolving criteria for procurement, including performance, cost efficiency, and supplier reliability. It highlights specific applications and case studies from companies like Topseed, UBTech, and others, and mentions industry events such as the 2026 World Robotics Conference.
Da Shen 1 is an industrial-grade reconfigurable four-finger dexterous hand jointly released by Heiman Technology and Leju Robotics. It was launched 27 days earlier than Boston Dynamics' four-finger hand. The hand features 11 active degrees of freedom, supports multiple configuration switches, and is suitable for tasks such as fine manipulation, large travel gripping, and wrapping grasping. Its design emphasizes structural stiffness, drive efficiency, and task adaptability, providing more flexible and efficient end-effector solutions for industrial applications.
This paper describes the development of intelligent robots for assembling GB300 tester trays, a complex manufacturing task requiring high precision and adaptability. The authors explore challenges such as uncertainty in part geometry and performance requirements, and present a technical strategy combining classical robotics, imitation learning, and simulation-based reinforcement learning. The work emphasizes the importance of real-world data and infrastructure for achieving reliable automation in industrial settings.
Researchers have developed an aquatic soft robot inspired by feather stars that can move in all three dimensions using only two actuators. The design utilizes mechanical intelligence to achieve efficient 3D maneuverability through three modes: jellyfish, fish, and rotor. The robot can explore underwater spaces, lift objects, and work collaboratively with other robots, with future plans to develop a fully wireless version.
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).