mlsanityData to 14 Oct 2026

Shadow Program Inversion with Differentiable Planning: A Framework for Unified Robot Program Parameter and Trajectory Optimization

arXiv 2409.08678 ↗First posted 13 Sep 2024cs.RO cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

This paper presents SPI-DP, a novel first-order optimizer capable of optimizing robot programs with respect to both high-level task objectives and motion-level constraints. To that end, we introduce DGPMP2-ND, a differentiable collision-free motion planner for serial N-DoF kinematics, and integrate it into an iterative, gradient-based optimization approach for generic, parameterized robot program representations. SPI-DP allows first-order optimization of planned trajectories and program parameters with respect to objectives such as cycle time or smoothness subject to e.g. collision constraints, while enabling humans to understand, modify or even certify the optimized programs. We provide a comprehensive evaluation on two practical household and industrial applications.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 26 of its 57 references that are in the map. First posted on arXiv 13 Sep 2024.

Builds on

26 of 57 references.

References · 26

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2015Automatic differentiation in machine learning: a surveycited by 855
  3. 2019DiffTaichi: Differentiable Programming for Physical Simulationcited by 258
  4. 2016Bayesian Optimization with Safety Constraints: Safe and Automatic Parameter Tuning in Roboticscited by 134
  5. 2016A Differentiable Physics Engine for Deep Learning in Roboticscited by 108
  6. 2020Scalable Differentiable Physics for Learning and Controlcited by 73
  7. 2017Continuous-Time Gaussian Process Motion Planning via Probabilistic Inferencecited by 49
  8. 2024The Elements of Differentiable Programmingcited by 43
  9. 2019Differentiable Gaussian Process Motion Planningcited by 28
  10. 2021Safe Pontryagin Differentiable Programmingcited by 26
  11. 2020Encoding Physical Constraints in Differentiable Newton-Euler Algorithmcited by 23
  12. 2024Practice Makes Perfect: Planning to Learn Skill Parameter Policiescited by 23
14 more
  1. 2020Multimodal Trajectory Optimization for Motion Planningcited by 12
  2. 2022Skill-based Multi-objective Reinforcement Learning of Industrial Robot Tasks with Planning and Knowledge Integrationcited by 10
  3. 2023Revisiting Implicit Differentiation for Learning Problems in Optimal Controlcited by 8
  4. 2022CALIPSO: A Differentiable Solver for Trajectory Optimization with Conic and Complementarity Constraintscited by 6
  5. 2021Robot Program Parameter Inference via Differentiable Shadow Program Inversioncited by 5
  6. 2021Trajectory Optimization with Optimization-Based Dynamicscited by 5
  7. 2020Metaheuristics "In the Large"cited by 4
  8. 2022Heuristic-free Optimization of Force-Controlled Robot Search Strategies in Stochastic Environmentscited by 4
  9. 2023Differentiable Task Assignment and Motion Planningcited by 4
  10. 2023STAMP: Differentiable Task and Motion Planning via Stein Variational Gradient Descentcited by 4
  11. 2020Local Policy Optimization for Trajectory-Centric Reinforcement Learningcited by 3
  12. 2023A Unifying Variational Framework for Gaussian Process Motion Planningcited by 2
  13. 2024MuTT: A Multimodal Trajectory Transformer for Robot Skillscited by 2
  14. 2023Meta-Learning Parameterized First-Order Optimizers using Differentiable Convex Optimizationcited by 1

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025AI-based Framework for Robust Model-Based Connector Mating in Robotic Wire Harness Installationcited by 0