RobKiNet: Robotic Kinematics Informed Neural Network for Optimal Robot Configuration Prediction
Yanlong Peng, Zhigang Wang, Yisheng Zhang, Pengxu Chang, Ziwen He,
Kai Gu, Hongshen Zhang, Ming Chen*
Yanlong Peng, Zhigang Wang, Yisheng Zhang, Pengxu Chang, Ziwen He,
Kai Gu, Hongshen Zhang, Ming Chen*
Task and Motion Planning (TAMP) is essential for robots to interact with the world and accomplish complex tasks. The TAMP problem involves a critical gap: exploring the robot's configuration parameters (such as chassis position and robotic arm joint angles) within continuous space to ensure that task-level global constraints are met while also enhancing the efficiency of subsequent motion planning. Existing methods still have significant room for improvement in terms of efficiency.
Figure. An overview of differentiable kinematics and its integration with neural networks. (a) Overview of the joint coordinate systems for the 6-DOF robotic arm kinematics. (b) The computational process for the analytical inverse kinematics solution. (c) The computational process for the analytical forward kinematics solution. (d) Non-differentiable operations and function modules within the program. (e) The motivation behind this study. (f) The basic concept of differentiable programming — ensuring the chain rule of the computational graph. This is achieved by transforming non-differentiable operators (red circles) into differentiable operators (green). There are other challenges to differential programming (2, 3), and solutions to all these problems will be detailed later. (g) The integration of IK with neural networks for chassis position prediction (CMP). (h) The integration of FK with neural networks for predicting both chassis position and joint angle configuration (FMP).
Recognizing that robot kinematics is a key factor in motion planning, we propose a framework called the Robot Kinematic Constraints Injection Network (RobKiNet).
RobKiNet integrates kinematic knowledge into neural networks to train models capable of efficient motion planning. We designed a Chassis Motion Predictor(CMP) and a Full Motion Predictor(FMP) using RobKiNet, which employed two entirely different sets of forward and inverse kinematics constraints to achieve loosely coupled control and whole-body control, respectively.
Experiments demonstrate that CMP and FMP can predict configuration parameters with 96.67% and 98% accuracy, respectively. That means that the corresponding motion planning can achieve a speedup of 24.24x and 153x compared to random sampling. Furthermore, RobKiNet demonstrates remarkable data efficiency. CMP only requires 1/71 and FMP only requires 1/15052 of the training data for the same prediction accuracy compared to other deep learning methods. These results demonstrate the great potential of RoboKiNet in robot applications.
Figure. Overview of this paper's issues, ideas, chapter relationships, and main contributions.
Real-world Experiments Demonstration
We deployed the CMP and the FMP for experiments in real-world scenarios.
The RobKiNet is utilized in the NeuroSymbolic TAMP framework and acts as the primitive "Approach" in guiding the entire autonomous movement of the AGV chassis. To comprehensively assess the performance of RobKiNet across various usage scenarios, we formulated four tasks. The corresponding task scenarios and requirements are illustrated in Figure . Through over 80 times of real-machine experiments, we observed that the RobKiNet accurately samples positions in actual disassembly scenarios with a 100% success rate.
Figure. Real-world experiments information and results.
Figure. Hardware platform
Contact Us
mingchen@sjtu.edu.cn
zhi.gang.wang@intel.com