AutoSpeed: Annotation-Free Stage-Adaptive Motion Speed Learning for Robot Manipulation

• URL: http://arxiv.org/abs/2607.01051v2
• Date: Sun, 05 Jul 2026 11:54:19 GMT
• Title: AutoSpeed: Annotation-Free Stage-Adaptive Motion Speed Learning for Robot Manipulation
• Abstract summary: IL-based visuomotor policies imitate the execution speed of expert demonstrations and operate with a fixed temporal prediction horizon.<n>We introduce AutoSpeed, a model-agnostic learning framework that enables existing visuomotor policies to predict trajectories with stage-adaptive motion speeds.<n>We show that AutoSpeed substantially reduces task execution time while also improving success rates.
• Score: 11.362770278564973
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Different stages of manipulation tasks exhibit varying levels of difficulty, suggesting stage-dependent motion speeds and temporal prediction horizons. However, existing IL-based visuomotor policies typically imitate the execution speed of expert demonstrations and operate with a fixed temporal prediction horizon, limiting flexibility and overall task throughput. In this paper, we introduce AutoSpeed, a model-agnostic learning framework that enables existing visuomotor policies to predict trajectories with stage-adaptive motion speeds, without requiring speed or stage annotations. We treat future trajectories at different speeds as candidate optimization targets, evaluate each candidate using a composite cost that trades off prediction error against prediction horizon, and optimize the policy toward the minimum-cost candidate. With a fixed-length action sequence, speed modulation adjusts the effective temporal prediction horizon: simple stages are executed faster with a longer prediction horizon, whereas complex stages are executed more slowly with a shorter prediction horizon. Specifically, we implement speed modulation in the frequency domain via the discrete cosine transform (DCT), which enables smooth, non-integer speed scaling and thus preserves motion continuity. Extensive evaluations show that AutoSpeed substantially reduces task execution time while also improving success rates. Under the AutoSpeed framework, the inferred motion speeds exhibit a strong correspondence with task stages.

Related papers

• Some Modifications to Our End-to-End UAV Planner [2.3190673374277524]
  The one-stage planner maps a single depth image and the robot state directly to a set of candidate trajectories.<n>We show that YOPO's multi-modal prediction span to distinct homotopy classes, treating each motion as a trajectory that confines the trajectory to a feasible - without explicit safe-building or construction penalties.
  arXiv  Detail & Related papers  (2026-08-16T13:39:46Z)
• SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning [44.701558695482476]
  We introduce SpeedTuning, a reinforcement learning framework specifically designed to enhance the speed of manipulation policies.<n>SpeedTuning learns to predict the optimal execution speed for actions, thereby complementing a base policy.<n>We provide empirical evidence that SpeedTuning achieves substantial improvements in execution speed, exceeding 2.4x speed-up.
  arXiv  Detail & Related papers  (2026-08-10T05:31:41Z)
• TempoVLA: Learning Speed-Controllable Vision-Language-Action Policies [58.40033352838586]
  Existing Vision-Language-Action models (VLAs) only inherit a single fixed speed from training demonstrations.<n>We observe that the magnitude of each predicted action already governs how fast the robot moves.<n>We turn this observation into TempoVLA, a single VLA whose execution speed is controlled by an explicit condition.
  arXiv  Detail & Related papers  (2026-06-04T17:59:40Z)
• Can Users Specify Driving Speed? Bench2Drive-Speed: Benchmark and Baselines for Desired-Speed Conditioned Autonomous Driving [50.09883181221131]
  We present Bench2Drive-Speed, a benchmark with metrics, dataset, and baselines for desired-speed conditioned autonomous driving.<n>We introduce explicit inputs of users' desired target-speed and overtake/follow instructions to driving policy models.<n>Our experiments show that, under proper re-annotation, models trained on regular driving data perform comparably to on expert demonstrations.
  arXiv  Detail & Related papers  (2026-03-26T17:26:41Z)
• FASTER: Rethinking Real-Time Flow VLAs [82.58822112377923]
  Real-time execution is crucial for deploying Vision-Language-Action (VLA) models in the physical world.<n>We show that reaction time follows a uniform distribution determined jointly by the Time to First Action (TTFA) and the execution horizon.<n>We propose Fast Action Sampling for ImmediaTE Reaction (FASTER) to overcome this issue.
  arXiv  Detail & Related papers  (2026-03-19T17:51:37Z)
• Learning Velocity and Acceleration: Self-Supervised Motion Consistency for Pedestrian Trajectory Prediction [16.532357621144342]
  We propose a self-supervised pedestrian trajectory prediction framework.<n>We leverage velocity and acceleration information to enhance position prediction.<n>We conduct experiments on the ETH-UCY and Stanford Drone datasets.
  arXiv  Detail & Related papers  (2025-03-31T16:17:45Z)
• Motion Forecasting in Continuous Driving [41.6423398623095]
  In autonomous driving, motion forecasting takes place repeatedly and continuously as the self-driving car moves. Existing forecasting methods process each driving scene within a certain range independently. We propose a novel motion forecasting framework for continuous driving, named RealMotion.
  arXiv  Detail & Related papers  (2024-10-08T13:04:57Z)
• AMP: Autoregressive Motion Prediction Revisited with Next Token Prediction for Autonomous Driving [59.94343412438211]
  We introduce the GPT style next token motion prediction into motion prediction. Different from language data which is composed of homogeneous units -words, the elements in the driving scene could have complex spatial-temporal and semantic relations. We propose to adopt three factorized attention modules with different neighbors for information aggregation and different position encoding styles to capture their relations.
  arXiv  Detail & Related papers  (2024-03-20T06:22:37Z)
• Motion Transformer with Global Intention Localization and Local Movement Refinement [103.75625476231401]
  Motion TRansformer (MTR) models motion prediction as the joint optimization of global intention localization and local movement refinement. MTR achieves state-of-the-art performance on both the marginal and joint motion prediction challenges.
  arXiv  Detail & Related papers  (2022-09-27T16:23:14Z)
• Motion Planning and Control for Multi Vehicle Autonomous Racing at High Speeds [100.61456258283245]
  This paper presents a multi-layer motion planning and control architecture for autonomous racing. The proposed solution has been applied on a Dallara AV-21 racecar and tested at oval race tracks achieving lateral accelerations up to 25 $m/s2$.
  arXiv  Detail & Related papers  (2022-07-22T15:16:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.