Apple's research team published the paper "Perceptive Humanoid Motion from Simulation" on arXiv, unveiling HALO, an autonomous control system for humanoid robots, and simultaneously open-sourcing the training and deployment code.
Boston Dynamics released a video showing an experiment in which the Atlas humanoid robot's feet were replaced with wheels for locomotion, combining classical model-based motor control with whole-body reinforcement learning strategies to maintain dynamic stability on unstructured real-world terrain.
Duatic has launched the DxTR general-purpose industrial robot, which uses a dual-arm plus mobile base structure and can grasp, carry, place, and drive, moving thousands of bins per day. The robot can lift 30 kg and carry 200 kg, and relies on sensing and backdrivability to collaborate safely with people. Both the software and the complete machine are developed and built in-house by Duatic in Switzerland.
The authors propose the first mmWave radar-based onboard perception-and-control dynamic obstacle avoidance system for small quadrotor UAVs. The system combines a lightweight interacting multiple models (IMM) tracker with a control barrier function (CBF) controller that directly outputs evasive accelerations, and derives sufficient avoidance conditions relating sensing range, relative speed, and control delay.
The paper uses an evolutionary algorithm to search for multicopter body morphologies suited to different objectives and constraints, and uses the non-dominated sorting genetic algorithm II (NSGA-II) to generate a set of bodies with performance trade-offs. Compared with a standard quadcopter, the Pareto front extrema improve maneuverability, thrust-to-weight ratio, and size by 489.2%, 23%, and 4.8%, respectively; the maneuverability improvement comes from motor tilt angles, and the thrust-to-weight ratio improvement comes from more propellers. Real drone racing experiments show that the evolved bodies outperform the standard body in number of gates passed, flight speed, and lap time.
LACART combines lookahead-based control barrier functions (CBFs), a relaxed control Lyapunov function (CLF), and quadratic programming (QP) to modify nominal vehicle commands in a unified optimization layer, handling both circular obstacles and track boundaries.
DKWBC proposes a dynamics-informed whole-body control framework that uses inverse-dynamics feedforward and joint PD feedback to supply tracking torques, while a residual policy learns the joint-torque corrections needed for balance. The authors state that this structure explicitly separates motion-related torque demands, simplifying the policy learning task, and that they validated in simulation and on real hardware that a single balancing policy can generalize across multiple motion datasets, maintaining tracking and balance under ground support, added payload, and physical interaction. The project homepage has been made public.
SUGARv2 proposes a unified framework for learning diverse visual humanoid loco-manipulation skills through interaction reference synthesis and a shared training pipeline. The method adapts reusable task templates to different spatial configurations to synthesize coordinated robot and object motion references at low cost, and uses reference-based imitation as a common learning objective across tasks, first training privileged teachers and then training visual policies through teacher distillation and reinforcement learning.
The authors propose TPS-A*, a lattice planner whose edges are kinematically feasible trajectories produced by parameterized trajectory generators, with certified collision checking performed in trajectory parameter space. A grid precomputed per footprint needs only a single pass over local obstacles to return a certified free distance for every candidate trajectory, making the per-node collision cost independent of footprint complexity and trajectory count.
The paper proposes a safety-critical control framework for nonholonomic multi-robot systems that lets each robot circumnavigate a stationary target at a different orbital radius while maintaining spacing from neighbors and avoiding collisions. Each robot measures only its distance to the target and exchanges information only with its two adjacent robots; it drives the radius to the target value via acceleration, balances the formation via turn-rate, and uses a higher-order control barrier function quadratic program (HOCBF-QP) to simultaneously constrain target distance, geofence, and inter-robot safety.
FastFunnels proposes a hierarchical multi-robot navigation method: a high-level reinforcement learning policy generates deformable elliptical envelopes in real time, while a low-level distributed model predictive control (DMPC) layer keeps robots within the envelopes and avoids collisions with each other.
FATHOM proposes a world-model-guided diffusion action-trajectory selection framework for relative tracking of autonomous underwater vehicles (AUVs) under sparse active light markers. The method first fuses marker observations, inertial measurements, action history, and temporal dynamics using a multimodal task-supervised recurrent state-space model to form a predictive tracking belief; a diffusion action prior then generates multiple multi-thruster trajectories, and the best candidate is selected under a receding-horizon objective and continuously updated based on onboard observations.
ForPC is a force-aware whole-body control framework that regulates a humanoid robot's cart-pushing speed through two-stage reinforcement learning. The method first infers force latent variables and state latent variables from onboard observations without external force sensors, then uses a residual policy to coordinate arm, waist, and leg movements; in simulation, cart-pushing speed RMSE is reduced by up to 68.2% compared with a locomotion-only control baseline, and lateral error on multi-curve routes is reduced by up to 74.2%.
The 18th installment of the Dexterous Hand Practice series discusses how to integrate impedance control and admittance control into the closed loop to avoid crushing objects during grasping. The content focuses on how these two control methods are integrated into the force-control closed loop of a dexterous hand.
The implementation path of dexterous hand force control from position loop and current loop to impedance control is systematically reviewed. The content is organized around these three control layers, explaining the role of each layer in dexterous hand force control and how they connect.
Apple's research team published the paper "Perceptive Humanoid Motion from Simulation" on arXiv, unveiling HALO, an autonomous humanoid robot control system, and validating it on the Unitree G1.
Why it matters: Apple's HALO splits task orchestration, motion generation, and tracking into three layers, and reports comparative results of 91% in unfamiliar environments and 70% on composite tasks, offering a look at its layered design and data synthesis pipeline.
Kinetix AI's SMASH high-dynamic humanoid table tennis system was exhibited at the Hang Seng Bank booth at HKFW 2026, where players from the Hang Seng Academy played matches against it. The original article did not disclose metrics such as the system's degrees of freedom, battery life, or match success rate.
Addressing the problem of step loss in open-loop stepper motors, this article presents a practical retrofit using encoder closed-loop control and stall detection. The material provides only the title and summary; it does not give a specific retrofit plan, hardware selection, or measured data.
SuperMimic proposes a humanoid robot motion-tracking framework that uses a differentiable world model with 32-step backpropagation rollouts to reduce training compute demands when the motion library is large. The framework models physical laws with a deterministic incremental model and uses real states as the starting point for imagined rollouts. The original is an in-depth analysis; the body provides only a summary.
Reinforcement learning policy optimization has evolved from REINFORCE and Actor-Critic to TRPO and then PPO, with the core goal of making policy gradient updates more stable and efficient. PPO uses clipping on the probability ratio between the new and old policies to limit the size of each update, balancing implementation complexity and stability, and is suitable for both continuous and discrete actions. The article also reviews GAE advantage estimation, the on-policy data flow, and tuning approaches for key hyperparameters such as learning rate and clip range.
Ubot's "Xingzhe Taishan" humanoid robot completed the official course as the only fully autonomous entrant in the first CMG World Robot Mount Tai Climbing Competition, taking second place. The entire climb was free of remote-control intervention, with the "brain" recognizing terrain and planning routes in real time and the "cerebellum" dynamically adjusting gait to handle uneven steps and slippery surfaces. Ubot says its commercial quadruped robot for inspection scenarios has carried out regular autonomous inspection in limited scenarios across multiple pilot projects in power and mining.
Apple released the HALO humanoid whole-body motion control system, which procedurally generates 1,000 virtual residential scenes and synthesizes nearly 300 hours of paired data, enabling the Unitree G1 to master skills such as sitting on chairs, turning sideways to pass through a half-open door, crouching under a table, and pushing boxes.
Why it matters: Apple's HALO uses simulation-synthesized data to train humanoid whole-body skills, transfers them zero-shot to real scenes, and ablation experiments show the contribution magnitude of each module.
The first OTA for the Qiyuan T1 robot will begin rolling out tomorrow, adding features such as autonomously recognizing stairs while following and switching to a quadruped form to go up and down stairs, and remote audio and video calls via the App. The update also adds Youdao HiEcho spoken-language practice, scheduled reminders, voice-initiated face and voiceprint enrollment, on-device video recording and AI image generation, plus new AI enlightenment courses and DIY capabilities including PrimeFLOW skins, PrimeHEART personality voices, and Prime Toolkit secondary-development configuration.
Yuance Future has released Anchor 1, a whole-body control model positioned as the System 0 cerebellum layer within a whole-body intelligence architecture. It uses five anchor points—the pelvis, left and right wrists, and left and right ankles—as commands and does not rely on online retargeting.
Why it matters: The original article provides quantitative comparisons of the whole-body control model's real-machine error and teleoperation collection efficiency, which can be used to assess how the cerebellum layer's precision affects the success rate of upper-layer VLA tasks.
Odometry calibration for a differential drive mobile chassis must address two key parameters: wheel diameter error and effective wheelbase, and the calibration effect must be verified through real-vehicle trajectories. The article analyzes these two sources of error, explaining their impact on odometry accuracy and the calibration approach.
Transferring motion capture data to a humanoid robot body requires handling skeleton alignment and scale factors, and setting up a final safety gate. The article centers on these three points, explaining how to adapt motion capture data to the joint and size differences of the humanoid body and add safety limits before execution.
Beijing Humanoid released the Hi-WBC general whole-body control model on the "Huisikaiwu" platform on November 2. The model combines human motion references with real physical constraints, enabling robots to adjust how they execute tasks as payload, external forces, and terrain change. It uses a continual skill expansion framework based on MoE and policy distillation, Contact-aware Learning that incorporates contact force prediction, and Perceptive Control that introduces terrain awareness.
Why it matters: The original text discloses Hi-WBC's technical approach to handling payload, external force, and terrain constraints beyond motion tracking, offering insight into how whole-body control is moving from demonstrations toward real work.
Engineers tested payload compensation on custom STS32XX firmware for the SO-ARM101/102; under a 250 g load, gravity compensation calibrated with an empty gripper reduced deviation from about 70 mm to about 35 mm. The team is seeking input on how to automatically adapt to different payloads, and the relevant code has been placed in the repository.
Centered on Cosserat rod mechanics, curvature traveling-wave actuation, and anisotropic friction, PyElastica uses three sets of simulations to demonstrate the jointless serpentine gliding mechanism of a continuum bionic snake. The article also examines the actual effect of superimposing a lifting wave on a fixed gait.
A quadrotor tracks a yellow target ball using only a 64×64 camera image. The official pretrained policy was loaded to complete tracking experiments in Isaac Lab, with the target visible about 40% of the time and the quadrotor autonomously hovering to reacquire the target after losing it. The article also discusses known gaps when migrating the policy to the DJI Tello.
A review proposes a functional "decoder-shaper-controller" framework for continuous myoelectric control of upper-limb exoskeletons: the decoder estimates intention-related variables from surface electromyography (sEMG), the shaper maps them to a control reference according to an assistance policy, and the controller realizes that reference through actuator control.
The authors propose an uncertainty-aware active exploration framework that can autonomously learn task-agnostic, generalizable dynamics models for soft robots. The method was evaluated on three simulation platforms—a continuum arm, an articulated fish in fluid, and a musculoskeletal leg—as well as on a real pneumatically actuated continuum arm. Compared with random exploration and task-specific baselines, it achieves higher modeling accuracy and yields higher returns in zero-shot control on unseen tasks, while remaining robust under sensing noise, actuation delays, and material nonlinearities.
The research team proposes HALO, a humanoid robot autonomy system that synthesizes vision-language-action data across roughly one thousand procedurally generated home scenes, eliminating reliance on teleoperation data collection. HALO learns skills including sitting on chairs, navigating to specified objects, stepping sideways through half-open doors, crouching and crawling under tables, and pushing boxes, and can combine them to complete long-horizon tasks. Without ever seeing real home scenes during training, HALO transfers zero-shot to the physical world, achieving a 91% success rate across 360 trials in six unmapped real-world environments.
Why it matters: The paper presents a complete path from synthesizing vision-language-action data using procedurally generated home scenes to zero-shot transfer to real robots; its approach to data scaling is worth referencing.
The research team learned an adiabatic spectral submanifold (aSSM) model of the full body of a trimmed-helicoid soft trunk robot from data and used it in model predictive control (MPC), completing a range of shape control tasks on hardware. In closed-loop dynamic shape regulation, the method outperformed existing linear baselines by a factor of 10; in human-in-the-loop interactive demos, it completed section-by-section shaping and balanced a tennis ball for more than 25 seconds, whereas the linear baselines failed to complete the task. A model-free reinforcement learning policy trained on hardware for 12 hours failed on the same balancing task.
The authors propose soccerbot, a hierarchical reinforcement learning framework that distills a structured latent motion prior from large-scale human soccer motion capture data, constraining policy learning within a dynamically feasible motion manifold. A single skill-conditioned policy thereby learns moving, dribbling, trapping, and kicking in a unified way, while maintaining human-like motion and fluid transitions during skill switching. After robustness-oriented simulation design, the framework transfers zero-shot to the Unitree G1 humanoid robot, completing the full soccer skill set in real-world environments.
The authors propose H^2MARL, a hierarchical heterogeneous multi-agent reinforcement learning framework for humanoid whole-body control. The method performs policy factorization with physical heterogeneity awareness at the joint degree of freedom (JDoF) level, conditioning each torque-actuated joint on descriptors of its own actuation, kinematic, and dynamic properties, and uses closed-loop policy-executor co-optimization to jointly adapt the high-level pose planner and low-level joint executor with shared interaction trajectories.
The study proposes a unified planning and control framework for mobile continuum robots, combining a global layer consisting of A* search and minimum-jerk planning with a local layer consisting of a geometry-consistent inverse kinematics solver and Riemannian optimization, to coordinate the mobile base and continuum deformation for obstacle avoidance. The authors compare it with multiple baseline planners in simulated pipe-traversal tasks and conduct real-world experiments on navigation among dense obstacles and manipulation in confined spaces. Results show improvements in motion consistency, feasibility, and adaptability.
This study examines the visual stabilization performance of boundary-layer sliding mode control (SMC) relative to causal baselines such as LPF and PID under plant-binding conditions, where disturbance velocity approaches or exceeds the actuator slew limit and the per-step actuation budget is constrained.
The paper proposes Future-Window Inverse Kinematics (FWIK), an online inverse kinematics selection method that considers several future waypoints, connects multiple IK candidates in a layered joint-space graph, ranks them with a motion-time estimate, and then uses a learned model to remove transition paths prone to self-collision.
This paper proposes a contact-implicit whole-body optimization framework for humanoid robots to generate dynamic motion trajectories on parameterizable complex terrains such as steps and ramps, without requiring a predefined contact sequence. The method approximates ground geometry with smooth surfaces and extends rigid-body interaction from planar surfaces to arbitrary smooth surfaces through new contact constraints, using nonlinear programming to plan motions across various terrain topologies. The authors conducted preliminary feasibility tests on non-coplanar terrains in simulation and on the real robot ErgoCub.