Live PLC visibility
Read designated monitored signals, gain tags, scheduling variables, and related process feedback from a ControlLogix system.
ALREADY IN USE
Built from real control-system work.
AI-supervised tuning and live parameter interaction are already being exercised with a Rockwell ControlLogix controller in MELD's helicopter autopilot development work. This is the practical foundation for bringing the same approach to other industrial control applications.
Initial integration is ControlLogix-focused. The specific gains, tuning authority, control laws, and operating limits are defined by each application—not by a universal preset. Loop setup, monitored-signal descriptions, and tunable-parameter mappings are editable so the same interface can support PI, PID, and custom control loops.
CAPABILITIES
Configure, supervise, validate.
Map a loop once, then supervise recommendations and record the measured response.
AI Supervised Adjustment
Change permitted PI and PID gains and custom tuning values during a test or commissioning session, using defined access and operating constraints.
Adaptive recommendations
Use response measurements and supervisory logic to help identify useful tuning changes for the specific system, including non-PID parameters such as lookahead, intercept, or scheduler values.
Recorded results
Capture the applied settings alongside the system response to support evaluation and repeatable tests.
TRY THE INTERACTIVE AI SUPERVISOR
One loop. Measured results. AI-supervised tuning.
Use the NAV lookahead example or switch to custom PI and PID loops. In Loop Setup, edit descriptions and PLC tags, then save the mapping. In AI Supervisor, analyze a sample, review a bounded recommendation, apply a simulated PLC request, and evaluate the response.
Demonstration values and AI recommendations are illustrative. The embedded application runs only in your browser and does not connect to or write to a PLC. Monitor / Recommend mode is also supported as a read-only workflow. This demo emphasizes AI Supervised Adjustment.
TWO DEPLOYMENT PATHS
Commissioning and flight / shop testing
Observe controller behavior, change authorized gain values, compare responses, and preserve a record of each test configuration.
Permanent supervisory integration
Apply the same technology to a deployed system with application-specific limits, access control, versioned settings, event logging, and defined fallback behavior.
Deployment boundary: The MELD interface does not replace the controller's safety logic or engineering validation. Permanent installations—and particularly aircraft installations—require system-specific testing and any applicable approvals. The website is not a PLC control endpoint.
ADAPTIVE GAIN CONTROL
Bring AI-supervised loop tuning and feedback recording to your ControlLogix system.
Describe the process, available gain tags, and how the system should be supervised.