Neural Motion Prediction for In-flight Uneven Object Catching
This addresses the challenge of real-time robotic catching of irregularly shaped objects, which is incremental as it builds on existing motion prediction methods with neural network enhancements.
The paper tackles the problem of predicting the motion of in-flight uneven objects for robotic catching by introducing a Neural Acceleration Estimator (NAE) and an end-to-end training method with a Differentiable Filter (NAE-DF), achieving success rates of 83.3% and 86.7% on specific objects in real-world tests.
In-flight objects capture is extremely challenging. The robot is required to complete trajectory prediction, interception position calculation and motion planning in sequence within tens of milliseconds. As in-flight uneven objects are affected by various kinds of forces, motion prediction is difficult for a time-varying acceleration. In order to compensate the system's non-linearity, we introduce the Neural Acceleration Estimator (NAE) that estimates the varying acceleration by observing a small fragment of previous deflected trajectory. Moreover, end-to-end training with Differantiable Filter (NAE-DF) gives a supervision for measurement uncertainty and further improves the prediction accuracy. Experimental results show that motion prediction with NAE and NAE-DF is superior to other methods and has a good generalization performance on unseen objects. We test our methods on a robot, performing velocity control in real world and respectively achieve 83.3% and 86.7% success rate on a ploy urethane banana and a gourd. We also release an object in-flight dataset containing 1,500 trajectorys for uneven objects.