True{\AE}dapt: Learning Smooth Online Trajectory Adaptation with Bounded Jerk, Acceleration and Velocity in Joint Space

TL;DR

True{ ext}dapt is a model-free, online trajectory adaptation method for robots that ensures smooth, bounded movements by predicting joint accelerations with neural networks, trained to balance a ball on a plate in simulation and transfer to real robots.

Contribution

It introduces a neural network-based, model-free approach for smooth online trajectory adaptation with bounded jerk, acceleration, and velocity in joint space.

Findings

Successfully balances a ball on a plate with a simulated KUKA iiwa robot.

Demonstrates direct transfer of the balancing policy from simulation to real robot.

Ensures smooth, bounded joint movements through adaptive neural network predictions.

Abstract

We present True{\AE}dapt, a model-free method to learn online adaptations of robot trajectories based on their effects on the environment. Given sensory feedback and future waypoints of the original trajectory, a neural network is trained to predict joint accelerations at regular intervals. The adapted trajectory is generated by linear interpolation of the predicted accelerations, leading to continuously differentiable joint velocities and positions. Bounded jerks, accelerations and velocities are guaranteed by calculating the range of valid accelerations at each decision step and clipping the network's output accordingly. A deviation penalty during the training process causes the adapted trajectory to follow the original one. Smooth movements are encouraged by penalizing high accelerations and jerks. We evaluate our approach by training a simulated KUKA iiwa robot to balance a ball on a plate while moving and demonstrate that the balancing policy can be directly transferred to a real robot.