# Learning to Throw Objects Safely in Multi-Obstacle Environments

arXiv: https://arxiv.org/abs/2607.06388 · HTML: https://mlsanity.com/paper/2607.06388/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 34 references that are in the map.
- First posted on arXiv 7 Jul 2026.

## Builds on (most cited first, 14 of 14)

- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Robot Operating System 2: Design, Architecture, and Uses In The Wild](https://mlsanity.com/paper/2211.07752/) (2022)
- [TossingBot: Learning to Throw Arbitrary Objects with Residual Physics](https://mlsanity.com/paper/1903.11239/) (2019)
- [Grasp Pose Detection in Point Clouds](https://mlsanity.com/paper/1706.09911/) (2017)
- [Self-supervised 6D Object Pose Estimation for Robot Manipulation](https://mlsanity.com/paper/1909.10159/) (2019)
- [Deep Learning Approaches to Grasp Synthesis: A Review](https://mlsanity.com/paper/2207.02556/) (2022)
- [Event-based Agile Object Catching with a Quadrupedal Robot](https://mlsanity.com/paper/2303.17479/) (2023)
- [MVGrasp: Real-Time Multi-View 3D Object Grasping in Highly Cluttered Environments](https://mlsanity.com/paper/2103.10997/) (2021)
- [Simultaneous Multi-View Object Recognition and Grasping in Open-Ended Domains](https://mlsanity.com/paper/2106.01866/) (2021)
- [Interactive Open-Ended Object, Affordance and Grasp Learning for Robotic Manipulation](https://mlsanity.com/paper/1904.02530/) (2019)
- [Throwing Objects into A Moving Basket While Avoiding Obstacles](https://mlsanity.com/paper/2210.00609/) (2022)
- [A Solution to Adaptive Mobile Manipulator Throwing](https://mlsanity.com/paper/2207.10629/) (2022)

## Abstract (arXiv)

Robotic throwing enables fast and efficient object placement beyond the robot's immediate workspace, but reliable throwing in cluttered environments remains underexplored. Existing approaches, such as TossingBot, learn throwing strategies from visual input but assume obstacle-free settings. In this paper, we address the problem of throwing objects into a target basket while avoiding obstacles placed randomly in the scene. We introduce a potential field state representation that compactly encodes both basket attraction and obstacle repulsion on a fixed-size grid, enabling reinforcement learning (RL) policies to generalize across arbitrary numbers and configurations of obstacles. The policy is initialized from kinesthetic demonstrations and optimized in simulation using three state-of-the-art RL algorithms (SAC, DDPG, TD3). Among these, SAC achieves the most consistent performance across scenarios. We compare the potential field representation against explicit state encodings and demonstrate that it achieves higher success rates and better scalability to unseen obstacle configurations. Real-robot experiments with unseen throwable objects confirm robust sim-to-real transfer, achieving up to 90% success in cluttered scenes. These results demonstrate that PFR provides a practical and robust representation for safe and efficient robotic throwing in unstructured environments. A video showcasing our experiments is available at: https://youtu.be/ZZnJf8ua2dE