End-to-end deep learning-based framework for path planning and collision checking: bin picking application

TL;DR

This work presents PPCNet, an end-to-end, imitation-learning–based framework for real-time path planning and collision checking in bin-picking. By coupling a planning network (modified MPNet) with a collision-checking network and training them via data aggregation (DAGGER), PPCNet achieves substantial reductions in planning time while maintaining competitive path lengths and success rates. The authors explore two collision-checker training strategies (binary vs. population-based labels) and demonstrate improvements through post-processing of training data (Binary State Contraction and resampling). Evaluations in simulation and on a Kinova Gen3 robot show PPCNet outperforms several state-of-the-art planners in speed and retains robust performance, highlighting its practical impact for industrial automation.

Abstract

Real-time and efficient path planning is critical for all robotic systems. In particular, it is of greater importance for industrial robots since the overall planning and execution time directly impact the cycle time and automation economics in production lines. While the problem may not be complex in static environments, classical approaches are inefficient in high-dimensional environments in terms of planning time and optimality. Collision checking poses another challenge in obtaining a real-time solution for path planning in complex environments. To address these issues, we propose an end-to-end learning-based framework viz., Path Planning and Collision checking Network (PPCNet). The PPCNet generates the path by computing waypoints sequentially using two networks: the first network generates a waypoint, and the second one determines whether the waypoint is on a collision-free segment of the path. The end-to-end training process is based on imitation learning that uses data aggregation from the experience of an expert planner to train the two networks, simultaneously. We utilize two approaches for training a network that efficiently approximates the exact geometrical collision checking function. Finally, the PPCNet is evaluated in two different simulation environments and a practical implementation on a robotic arm for a bin-picking application. Compared to the state-of-the-art path planning methods, our results show significant improvement in performance by greatly reducing the planning time with comparable success rates and path lengths.

Figures (6)

• Figure 1: Path planning for pick and place operation.
• Figure 2: Path Planning and collision checking network (PPCNet).
• Figure 3: Post processing procedure: a) The generated path by the planner, b) Path after binary state contraction, c) Path after resampling.
• Figure 4: End-end-training process of PPCNet: Top row: The imitation learning and data aggregation processes for training the planner network. Bottom row: Population-based probability estimation and collision checker network training processes.
• Figure 5: Experimental environments: a) UR5 scene, b) UR5 scene with a wall as an obstacle, c) Real-world implementation on Kinova Gen3 .