Bardo.Examples.Benchmarks.Dpb.DpbWoDamping (Bardo v0.1.0)

View Source

Double Pole Balancing without Damping.

This variant of the DPB task does not provide velocity information, making it a harder task. The neural network needs to develop recurrent connections to compute velocity estimates.

Summary

Functions

get_phys_config(owner, cortex_id, scape_id)

Returns the physical configuration for sensors and actuators required by the DPB without damping experiment.

neuron_pattern(owner, agent_id, cortex_id, substrate)

Creates a neuron pattern for the DPB without damping experiment.

Functions

get_phys_config(owner, cortex_id, scape_id)

Returns the physical configuration for sensors and actuators required by the DPB without damping experiment.

Parameters

  • owner - Owner ID
  • cortex_id - Cortex ID
  • scape_id - Scape ID

Returns

  • Map with sensor and actuator definitions

neuron_pattern(owner, agent_id, cortex_id, substrate)

Creates a neuron pattern for the DPB without damping experiment.

Parameters

  • owner - Owner ID
  • agent_id - Agent ID
  • cortex_id - Cortex ID
  • _substrate - Substrate (unused in this implementation)

Returns

  • Neuron pattern with mappings between sensors/actuators and neurons