📩 Aperture@thegaogroup.com
Distributed Intelligence, Robotics and Physical Logistics
Physical AI and AIoT Research Team of
Aperture Venture Studio, GAO RFID Inc. and GAO Tek Inc.
This program extends individual AIoT systems into coordinated operations involving multiple devices, locations or machines.
Resilient Edge AI and Distributed Industrial Intelligence
Develop systems that distribute inference and decision-making between sensors, gateways, local computers and cloud services.
The core problem is deciding what must happen locally and what can depend on remote resources, especially when communication or computing capacity is limited.
Research questions
How should models and tasks be allocated across devices? What capabilities should continue during an outage? How can devices recover without acting on outdated information?
Initial demonstration
Several edge nodes monitoring and controlling a small process while network faults are introduced.
Evaluation
Latency, bandwidth, compute and energy use, continued service during outages and recovery time.
Multi-Agent Coordination for Industrial Operations
Study how specialized agents coordinate inspection, maintenance, inventory, production and material movement.
Each agent should have defined responsibilities, permissions and shared-state rules. Research should address conflicts and uncertainty rather than merely placing several conversational agents in a workflow.
Research questions
How can agents resolve competing demands for equipment or materials? When does decentralized coordination outperform a central scheduler?
Initial demonstration
Coordinate material replenishment, inspection and maintenance requests across several simulated or physical stations.
Evaluation
Completion time, resource utilization, conflicting actions, deadlocks and robustness to agent failure.
AI-Assisted Material Handling, Warehouse Robotics and Physical Supply-Chain Execution
Develop intelligence for task assignment, replenishment, routing and exception handling using existing robots or material-handling equipment.
This combines warehouse robotics, autonomous material handling and the physical-execution portion of supply-chain optimization. RFID and UWB can provide asset context, while robot systems perform movement.
Research questions
How should uncertain inventory and location information affect dispatch decisions? How can the system replan when routes, loads or destinations change?
Initial demonstration
A robot or mobile platform moves identified materials between defined stations.
Evaluation
Delivery accuracy, throughput, travel distance, congestion, intervention rate and recovery from failed transfers.
Adaptive Robotic Assembly and Machine-Tool Assistance
Investigate how robots and machine tools adapt to variation in components, positioning, tool condition and process state.
Narrow the work to one task, such as insertion, fastening, machine loading or adjustment of a permitted machining parameter. Access to equipment and domain expertise should precede project approval.
Research questions
Can vision and force sensing support adaptation to part variation? Can a model recognize a failed operation and choose an appropriate recovery?
Initial demonstration
A repeatable assembly or machine-tending task with deliberately varied conditions.
Evaluation
Success rate, cycle time, dimensional or assembly quality, setup effort and recovery success.