dPMP-Deep Probabilistic Motion Planning: A use case in Strawberry Picking Robot

TL;DR

This paper introduces a deep probabilistic extension of movement primitives for robot learning from demonstrations, specifically applied to strawberry harvesting, improving trajectory accuracy and enabling trajectory sampling for secondary objectives.

Contribution

It presents a novel deep probabilistic model extending deterministic DMPs, along with a new training method for domain-specific latent features, enhancing robot trajectory prediction and flexibility.

Findings

Probabilistic model outperforms existing methods in trajectory accuracy.

Latent space learning significantly improves prediction performance.

Sampling from the distribution enables trajectory optimization for secondary objectives.

Abstract

This paper presents a novel probabilistic approach to deep robot learning from demonstrations (LfD). Deep movement primitives (DMPs) are deterministic LfD model that maps visual information directly into a robot trajectory. This paper extends DMPs and presents a deep probabilistic model that maps the visual information into a distribution of effective robot trajectories. The architecture that leads to the highest level of trajectory accuracy is presented and compared with the existing methods. Moreover, this paper introduces a novel training method for learning domain-specific latent features. We show the superiority of the proposed probabilistic approach and novel latent space learning in the lab's real-robot task of strawberry harvesting. The experimental results demonstrate that latent space learning can significantly improve model prediction performances. The proposed approach allows to sample trajectories from distribution and optimises the robot trajectory to meet a secondary objective, e.g. collision avoidance.