A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms
By Oliver Kroemer, Scott Niekum, George Dimitri Konidaris. 2019
509 citations recorded by the source corpus.
A key challenge in intelligent robotics is creating robots that are capable of directly interacting with the world around them to achieve their goals. The last decade has seen substantial growth in research on the problem of robot manipulation, which aims to exploit the increasing availability of affordable robot arms and grippers to create robots capable of directly interacting with the world to achieve their goals. Learning will be central to such autonomous systems, as the real world contains too much variation for a robot to expect to have an accurate model of its environment, the objects in it, or the skills required to manipulate them, in advance. We aim to survey a representative subset of that research which uses machine learning for manipulation. We describe a formalization of the robot manipulation learning problem that synthesizes existing research into a single coherent framework and highlight the many remaining research opportunities and challenges.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Control and Systems Engineering
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sergey Levine
- Chelsea Finn
- Pieter Abbeel
- Yuke Zhu
- Li Fei-Fei
- Trevor Darrell
- Timothy M. Hospedales
- Amos J. Storkey
In Computer Vision and Pattern Recognition
- Sergey Levine
- Luc Van Gool
- Dacheng Tao
- Li Fei-Fei
- Trevor Darrell
- Chelsea Finn
- Jitendra Malik
- Bernt Schiele
In Control and Systems Engineering
- Sergey Levine
- Chelsea Finn
- Yuke Zhu
- Pieter Abbeel
- Ajay Mandlekar
- Mrinal Kalakrishnan
- Li Fei-Fei
- Julian Ibarz
Citations in the corpus
- robosuite: A Modular Simulation Framework and Benchmark for Robot Learning (2020, methodology): “Data-driven algorithms [9], such as reinforcement learning [13, 7] and imitation learning [12], provide a powerful and generic tool in robotics.”
- Self-Supervised Correspondence in Visuomotor Policy Learning (2019, background): “An overview of the broader topic of robot learning in manipulation is provided in [8].”
- Meta-Learning in Neural Networks: A Survey (2020, background): “Knowledge transfer in robotics is often best studied compositionally [231].”
- How to train your robot with deep reinforcement learning: lessons we have learned (2021, background): “More recently, Kroemer et al. (2019) reviewed the learning algorithms for manipulation tasks.”
- “Good Robot!”: Efficient Reinforcement Learning for Multi-Step Visual Tasks with Sim to Real Transfer (2019, background): “While these types of constraints are intuitive, incorporating them into Deep RL (DRL) in a manner that leads to reliable and efficient learning is nontrivial [1], [2].”
- Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes (2024, background): “Moreover, many of these real-world successes are demonstrated on short-horizon tasks; further work is warranted to build agents that can reason over longer periods of time and compose learned abilities together to solve long-horizon tasks (11, 123, 132, 150, 151).”