Motivation

Anticipation is a key ability for advanced autonomous systems, especially if they operate in densely crowded environments and alongside humans. Examples of tasks benefiting from predictive human motion models include robot task and motion planning, automated driving, physical human-robot collaboration, intelligent video surveillance. Foreseeing how a scene involving multiple agents evolves over time and incorporating predictions in a proactive manner allows for novel ways of planning, active perception, model predictive control, or human- robot interaction. The growth of community interest to the area in recent years is evident, in particular from the publication trends in regular paper sessions of the recent robotics conferences (ICRA, IROS, RSS and others). Furthermore, the area is recognized as a major component/research direction by several big players in the automotive industry.

The proposal of this Special Issue is motivated by the success of the first workshop on “Long-term Human Motion Prediction”, which took place at ICRA 2019 in Montreal, see https://motionpredictionicra2019.github.io/. The event has received a high number of attendees (more than 100), and the audience, including renowned experts in motion prediction, has shared their positive feedback to the quality of the talks and the submitted extended abstracts.

The special issue aims to collect the several contributions presented during the workshop and to further welcome possible submissions from the interested researchers by launching a call-for-papers.

Topical Areas

Topics of interest for this special issue include and are not limited to:

• Motion trajectory prediction in 2D and 3D
• Predicting articulated human motion
• Early action and activity recognition
• Visual scene prediction
• Predictive planning and control
• Anticipation of group and crowd motion
• Human motion prediction and safety
• Human-Robot Interaction considering predictions
• Evaluation of prediction algorithms: datasets, metrics and benchmarks
• Applications of motion prediction techniques

Timeline

The special issue will follow the following timeline: