In this research, we introduce a novel approach to the challenge of suction grasp point detection. Our method, exploiting the strengths of physics-based simulation and data-driven modeling, accounts for object dynamics during the grasping process, markedly enhancing the robot's capability to handle previously unseen objects and scenarios in real-world settings. We benchmark DYNAMO-GRASP against established approaches via comprehensive evaluations in both simulated and real-world environments. DYNAMO-GRASP delivers improved grasping performance with greater consistency in both simulated and real-world settings. Remarkably, in real-world tests with challenging scenarios, our method demonstrates a success rate improvement of up to 48% over SOTA methods. Demonstrating a strong ability to adapt to complex and unexpected object dynamics, our method offers robust generalization to real-world challenges. The results of this research set the stage for more reliable and resilient robotic manipulation in intricate real-world situations.
Overall Method
Overall Method
Single Pick on Isaac GYM
Single Pick on Isaac GYM
Parallelized environments on Isaac GYM
Parallelized environments on Isaac GYM
Random configurations generated in Isaac GYM:
Random configurations generated in Isaac GYM:
Real-world Experiments: Adversarial Set
Real-world Experiments: Adversarial Set
Real-world adversarial evaluation: DYNAMO GRASP (our method), DexNet, centroid-based, and a DYNAMO+DexNet hybrid. The color-coded points are examples of the suggested grasp points from various methods.
Our Method
Baseline#1
Baseline#2
Extra Experiment
Real-world Experiments: Common Set (T - target object)
Real-world Experiments: Common Set (T - target object)
Real-world Experiments: Challenging Set (T - target object)
Real-world Experiments: Challenging Set (T - target object)
