ICAPS 2026 Descriptions of Special Tracks

ICAPS-26 has an explicitly broad interpretation of planning and scheduling. To highlight certain areas of interest and ensure that papers in those areas are reviewed appropriately, ICAPS-26 has four special tracks. If your work does not fit within one of these special tracks, please submit to the default `main track’. If you are unsure whether your work should be submitted to a special track or the main track, feel free to contact the chairs of the special track.

Track: Human-aware Planning and Scheduling (Chairs: Reuth Mirsky and William Yeoh)

The human-aware planning and scheduling track welcomes papers that address any challenges inherent to settings where a person needs to use, interact, collaborate or work along with planning and scheduling systems. The goal of this track is to foster advances in algorithms, design and modeling that enable planning and scheduling systems to interact effectively, transparently, and safely with people. We welcome submissions on novel algorithms, modeling techniques, challenges, and case studies of fielded systems. The track will also accept papers that address these challenges using Reinforcement learning or other learning mechanisms for sequential decision making.

Topics of interest include, but are not limited to:

Human-aware planning, scheduling, and execution
Representing, modeling, and reasoning about human goals, plans, preferences, intent, capabilities, and constraints
Mixed-initiative planning and scheduling with humans in the loop
Learning from demonstrations, feedback, or preference signals for planning in the presence of humans
Explainable, interpretable, and predictable planning & scheduling behavior
Trust, safety, ethics, transparency, fairness, communication and cooperation in human-aware planning systems
Interfaces, visualization, and methods for effective human–planner communication
Applications in human–robot teaming, collaborative decision support, assistive systems, and real-world scheduling scenarios

Track: Industry and Applications (Chairs: Jeremy Frank and Neil Yorke-Smith)

The Industry and Applications track (formerly Novel Applications track) invites papers that demonstrate how planning and scheduling methods are applied to solve problems in industry and commerce, government, or emerging societal contexts. This track aims to bring scientific results from planning and scheduling to practice, and also aims to identify scientific challenges by analyzing the limitations of current results to tackle real-world problems. We solicit submissions and participation from the private and public sectors, as well as from academics working on applications. We encourage submissions to provide rigorous evaluation using real-world data, and if applicable, to discuss the deployment of the methods and systems.

Topics of interest include, but not limited to:

Description and modelling of novel application domains
Engineering issues in using, deploying and scaling planning and scheduling techniques
Integration of multiple planning and scheduling techniques, or of planning and scheduling techniques with techniques from other areas or disciplines
User interface design, visualization, or human-system collaboration for a planning and scheduling application
Experiences in development, deployment, and maintenance of planning and scheduling applications
Evaluation, testing, and validation of planning and scheduling applications in societal or industrial environments
Assessment of impact on end users, customers, markets, or society at large
Industry / application challenge problems in planning and scheduling (including benchmark instances)

Track: Planning and Learning (Chairs: Blai Bonet and Akshat Kumar)

The Planning and Learning track welcomes submissions at the intersection of planning and scheduling, machine learning and reinforcement learning. We particularly welcome submissions that provide rigorous empirical evaluation or theoretical insights into the interaction between learning, planning and scheduling fields to improve adaptability and scalability of sequential decision-making systems. Submissions must clearly make a case for their relevance to automated planning and scheduling systems.

Topics of interest include, but not limited to:

Learning heuristics, domain and action models, or other forms of control knowledge for planning
Learning generalized policies, sketches, goal decompositions, landmarks for planning
Neuro-symbolic approaches for representing and reasoning with learned models
Theoretical aspects of planning and learning
Multi-agent planning and learning
Planning and scheduling applied to machine learning systems
Learning to abstract or decompose problem structure for effective planning
Planning and scheduling methods to improve the performance of foundation models (LLM, VLM, VLA, etc.)
Planning and scheduling using foundation models (LLM, VLM, VLA, etc.)
Model-free approaches to planning and scheduling built from data

Track: Robotics (Chairs: Federico Pecora and Alan Fern)

The Robotics track welcomes research contributions in the general areas of planning and scheduling, which relate to the advancement of intelligent robots. The track aims to bridge high-level planning and scheduling with low-level control to enable robust, adaptive, and intelligent robot behavior. We especially encourage submissions that demonstrate and evaluate their approach on physically-realistic simulation platforms such as Gazebo and IsaacSim, and in physical robots, where possible. Submissions must demonstrate tight coupling between planning algorithms and robotic perception, control, or learning.

Topics of interest include, but not limited to:

Task and motion planning
Learning methods for robot planning
Safety, explainability, legibility, and ethics in robot planning
Planning and learning for human-robot teaming and interaction
Formal methods for robot planning and control
Planning for perception and SLAM
Plan execution, failure detection, and recovery in robotic tasks
Multi-robot planning, scheduling/coordination, and execution
Mixed initiative planning and adjustable autonomy
Temporal, hierarchical, continuous-space planning for embodied agents
Benchmarking planning domains for robotics
Integrating planning with foundation models (LLM, VLM, VLA, etc.) for robotics