Plan4ARI
Integrated architecture for trustworthy industrial autonomy
Develop an Artificial Robotic Intelligence stack that makes industrial robots easier to program, safer to deploy, and more adaptive in dynamic human-robot environments.
Partners
Industrial robotics and applied research, side by side
Funding
Co-funded by MUR through the Italian Fund for Applied Sciences, FISA 2024.
_2024.svg-231702567.png){kind=logo}
Project at a glance
Industrial autonomy, planned for trust.
Industrial robots remain constrained by specialised programming, long engineering cycles and limited adaptation to changing environments. Plan4ARI integrates natural communication, task planning, sensing, motion planning, real-time control and human-centred evaluation.
- Duration
- 48 months
- Area
- Advanced Manufacturing
- Host Institution
- COMAU
- Research Organization
- CNR
- Principal Investigator
- Nicola Pedrocchi
From AGI to ARI
Robust, safe and verifiable intelligence, embedded in industrial robot workflows.
Plan4ARI does not claim unrestricted general intelligence. It builds Artificial Robotic Intelligence across four levels:
- Programming intelligence
- Task intelligence
- Motion intelligence
- Execution intelligence
Architecture
Two cooperating layers for adaptive robot intelligence.
- Open IPC cognition, world modelling and planning.
- Core IPC deterministic execution, safety and real-time control.
Research modules
Four work streams shape the Plan4ARI stack.
- M1 Open IPC Low-Code Programming and Natural Communication Transforms operator intent into verifiable task representations for industrial robot workflows. Explore the module
- M2 Open IPC Intelligent Sensing and Safe Motion Planning Connects scene understanding, feasibility checks and motion planning for adaptive execution. Explore the module
- M3 Core IPC Neural Motion Control Explores learned control and adaptation while preserving industrial safety constraints. Explore the module
- M4 Across both layers Trustworthiness and Human-Centric Optimization Frames evaluation around explainability, acceptance, safety and operator confidence. Explore the module
Demonstrators
Representative industrial scenarios for adaptive autonomy.
-
01
Warehouse Picking
Heterogeneous objects, natural-language commands, task allocation and online replanning.
-
02
Remanufacturing
Disassembly of mechanical components with flexible human intervention and force-sensitive interaction.
-
03
Ship Welding
Mobile manipulators in large, cluttered ship structures, with nearby operators and runtime safety.
Roadmap
Four phases over 48 months
-
Months 1-12
Scientific-plan updates, benchmarking and first mockups.
MS1 · M12
-
Months 13-24
Planning chain, sensing fusion, motion planning and neural-controller prototypes.
MS2 · M24
-
Months 25-36
Integration, online replanning and evaluation campaigns.
MS3 · M36
-
Months 37-48
Integrated demonstrators, exploitation and project closure.
MS4 · M48
News
Latest updates
-
Six months in: technical meeting on Activities A2–A6
The first six months have flown by. At the July technical meeting, the teams reviewed progress across Activities A2–A6 as the scientific plan update nears completion.
-
A new challenge from COMAU: something big is coming
COMAU has put a new challenge on the table: new technologies for the Plan4ARI demonstrators. We can’t reveal them yet, but stay tuned.
-
Handshakes at COMAU: the physical kick-off
After weeks of video calls, the Plan4ARI teams finally met in person at COMAU for a full day of presentations, discussions and, yes, handshakes.
-
Rolling up our sleeves: the first technical scoping meeting
Eleven days after the kick-off, the teams moved from vision to work plan, setting the first technical steps for each research module.
-
Plan4ARI is officially under way: our virtual kick-off
Two teams, one ambition: on 13 February COMAU and CNR met online to launch Plan4ARI, a 48-month journey towards robots that are easier to program, safer to deploy…
Contact
Interested in collaborating?
Write to the research and industrial leads of Plan4ARI about the project, the demonstrators or possible collaborations.
-
Nicola PedrocchiPrincipal Investigator, CNR-STIIMA
-
Nicola LongoProject manager, COMAU