4.0 Env - Dogfight (WIP)

.gitignore

@@ -7,6 +7,13 @@ logs/
 # Build dir
 build/
 
+# Dogfight test binaries (compiled from .c via gcc one-liners)
+ocean/dogfight/test_flight_dynamics
+ocean/dogfight/tests/test_*
+!ocean/dogfight/tests/test_*.c
+!ocean/dogfight/tests/test_*.h
+!ocean/dogfight/tests/test_*.py
+
 # hipified cuda extensions dir [HIP/ROCM]
 pufferlib/extensions/hip/
 

config/dogfight.ini

@@ -0,0 +1,51 @@
+[base]
+env_name = dogfight
+
+[vec]
+total_agents = 4096
+num_buffers = 8
+num_threads = 8
+
+[policy]
+num_layers = 2
+hidden_size = 128
+
+[env]
+num_agents = 1
+max_steps = 300
+obs_scheme = 1
+curriculum_enabled = 0
+curriculum_randomize = 0
+reward_aim_scale = 0.001695
+reward_closing_scale = 0.0001
+penalty_neg_g = 0.035
+control_rate_penalty = 0.002
+low_altitude_threshold = 1200.0
+low_altitude_penalty = 0.005
+speed_min = 50.0
+aim_decay_stage = 30.0
+shaping_decay_start = 100_000_000
+shaping_decay_end = 150_000_000
+energy_gain_scale = 0.001
+energy_loss_scale = 0.0005
+energy_advantage_scale = 0.004
+
+[train]
+total_timesteps = 600_000_000
+learning_rate = 0.00045
+minibatch_size = 65536
+gamma = 0.99
+gae_lambda = 0.999
+ent_coef = 0.0024
+clip_coef = 0.11
+vf_coef = 2.9
+vf_clip_coef = 1.5
+max_grad_norm = 1.8
+beta1 = 0.9896
+beta2 = 0.94
+eps = 1.49e-08
+prio_alpha = 0.99
+prio_beta0 = 0.99
+vtrace_rho_clip = 2.79
+vtrace_c_clip = 3.5
+replay_ratio = 1.0