New models connect internet knowledge to robot movement
Researchers tracked the development of vision-language-action models from 2022 to 2026.
Researchers tracked the development of vision-language-action models from 2022 to 2026.
A system called SALA aligns human movements with robot actions using language descriptions.
A new method called InneR helps robots respond consistently to changes in their environment.
A new architecture evaluates ongoing information acquisition relative to declared task requirements.
Researchers developed a method to help robots learn how to move and balance using a mathematical approach.
The SWAM model uses large amounts of video and demonstration data to improve how robots perform tasks.
The SLA model uses continual pre-training to improve how robots handle tasks in three-dimensional space.
A two-stage framework reuses control knowledge to learn visual policies from limited demonstrations.
This framework synthesizes full manipulation trajectories from unseen robot base poses including dynamic object interactions.
A method uses sparse anchors to register a complete trajectory to a reference map.
ForceLDM uses force and motion data to help robots manage contact-rich manipulation.
Researchers integrated low-thrust propulsion dynamics into a planning framework for on-orbit servicing.
The BURGER benchmark uses a scalable pipeline to test robotic drawing tasks in simulation and the real world.
The DAWN framework uses multi-modal sensory integration to perform unfastening through two sequential stages.
Fourteen methods were compared using five different quadrupeds on various 3D terrains.
SwapBot uses low-cost adapters to turn household tools into robot parts for tasks like cooking.
Researchers tested how vision models use egocentric video to track players and balls in tennis.
VPR-Space evaluates visual place recognition capabilities on extraterrestrial surfaces.
The DexPort interface uses hand hardware to record motions for dexterous manipulation tasks.
Echo combines task status signal learning and action generation within a single Vision Language Action (VLA) model.
A protocol called WarpWeaver uses negotiation to help robots reach near-optimal agreements.
This system uses probabilistic point-level confidence updating to improve 3D semantic maps.
Post-hoc kernel-variance heads estimate uncertainty from a single policy.
The A2I-Calib++ framework helps multi-IMU systems on legged robots work more accurately.