added Phase-2 code

Phase_2_Control_Toy_Car_Continuous_Env/DDQN_final_trained_network/configs.txt

@@ -0,0 +1,79 @@
+############################
+DDQN-CONFIGS AND HYPERPARAMS
+############################
+
+# define all the configurations and parameters here
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+learning_rate = 5e-4
+buffer_size = int(1e5)
+batch_size = 64
+discount_factor = 0.99
+update_target_freq = 5
+tau = 1e-3
+epsilon_start = 0.2
+epsilon_end = 0.0002
+epsilon_decay_rate = 1e-6
+hidden_dim1 = 64
+hidden_dim2 = 64
+max_t = 6000 # maximum number of timesteps per episode
+max_episodes = 300 # maximum number of episodes to train the agent
+save_logs_freq = 25 # save logs after every save_logs_freq episodes
+save_model_ckpts_freq = 25 # save model checkpoints after every save_model_ckpts_freq episodes
+train_stats_print_freq = 5 # print training statistics after every train_stats_print_freq episodes
+save_model_path = './train_logs/model_checkpoints/'
+save_logs_path = './train_logs/logs/'
+
+
+
+######################
+ENVIRONMENT-CONFIGS
+######################
+
+# Circular path parameters
+self.center = (self.window_size[0] // 2, self.window_size[1] // 2 - 50)
+self.radius = 380
+self.path_thickness = 80
+self.no_path_radius = 180
+self.no_path_outer_radius = self.radius + self.path_thickness + 20
+
+# Stochastic wind parameters
+self.wind_mean_x = 0.3
+self.wind_variance_x = 0.05
+self.wind_speed_x = np.random.normal(self.wind_mean_x, self.wind_variance_x)
+
+self.wind_mean_y = 0.3
+self.wind_variance_y = 0.05
+self.wind_speed_y = np.random.normal(self.wind_mean_y, self.wind_variance_y)
+
+# Car parameters
+self.car_length = 30
+self.car_width = 20
+self.car_speed = 1.2
+self.car_angle = math.pi / 2  # Start facing upward
+self.car_x, self.car_y = self.center[0] + self.radius - (self.path_thickness//2), self.center[1]
+self.start_pos_x, self.start_pos_y = self.car_x, self.car_y
+self.half_lap = False
+self.three_quarter_lap = False
+
+# Radar parameters
+self.num_rays = 5
+self.ray_lengths = np.array([50, 60, 70, 60, 50])
+self.ray_angles = np.array([-math.pi / 4, -math.pi / 8, 0, math.pi / 8, math.pi / 4])
+self.ray_values = np.array([0] * self.num_rays)  # Store radar ray values (1, -1, or 0)
+self.ray_x = np.array([0] * self.num_rays)
+self.ray_y = np.array([0] * self.num_rays)
+
+# State parameters and action space
+self.state_dim = self.num_rays + 3
+self.action_dim = 3
+
+# Rewards and score
+self.reward_inside_path = 2
+self.reward_outside_path = -15
+self.score = 0
+self.time_penalty = -1
+self.angle_change_penalty = -10
+self.goal_reward = 5000
+self.three_quarter_lap_reward = 1500
+self.no_access_hit_penalty = -200
+self.obstacle_sense_penalty = -5

Phase_2_Control_Toy_Car_Continuous_Env/DDQN_final_trained_network/loss_logs.csv

@@ -0,0 +1,301 @@
+Episode,Loss
+0,46.97676086425781
+1,350.78729248046875
+2,246.4852294921875
+3,89.11424255371094
+4,35061.25
+5,755.3390502929688
+6,1053.97509765625
+7,1186.527099609375
+8,182.76678466796875
+9,617.65234375
+10,968.5218505859375
+11,2013.9727783203125
+12,276.64483642578125
+13,1139.0655517578125
+14,784.6294555664062
+15,705.278076171875
+16,1086.9947509765625
+17,236.15655517578125
+18,210.43490600585938
+19,582.9960327148438
+20,1660.1112060546875
+21,285.4600830078125
+22,1001.1182861328125
+23,658.727783203125
+24,470.072265625
+25,1727.394775390625
+26,1044.6287841796875
+27,336.47442626953125
+28,2500.18310546875
+29,1244.27734375
+30,2103.3955078125
+31,819.7290649414062
+32,875.4436645507812
+33,1124.598388671875
+34,1243.79052734375
+35,1880.600341796875
+36,289.2767333984375
+37,2439.1484375
+38,401.39886474609375
+39,194.1799774169922
+40,1163.352783203125
+41,505.58441162109375
+42,386471.65625
+43,206.86944580078125
+44,342.10968017578125
+45,1325.816650390625
+46,410.22174072265625
+47,1020.5517578125
+48,605.8984985351562
+49,1357.3321533203125
+50,1886.571044921875
+51,210.90013122558594
+52,211.42311096191406
+53,203.08203125
+54,682.6439208984375
+55,1682.65380859375
+56,192.31179809570312
+57,353.06134033203125
+58,902.114501953125
+59,639.5557250976562
+60,655.00537109375
+61,329.2009582519531
+62,187.648193359375
+63,108.52923583984375
+64,113.84822082519531
+65,798.8270263671875
+66,213.05551147460938
+67,1095.076171875
+68,317.6368408203125
+69,438.614501953125
+70,57.4990234375
+71,380.6076354980469
+72,66.27198028564453
+73,1677.48828125
+74,2813.37646484375
+75,471.6524658203125
+76,1249.3409423828125
+77,229.22091674804688
+78,86.7664794921875
+79,685.968994140625
+80,1340.06201171875
+81,89.82808685302734
+82,275.87225341796875
+83,1064.884033203125
+84,213.00741577148438
+85,740.108154296875
+86,410.486328125
+87,1077.50341796875
+88,152.51947021484375
+89,885.6062622070312
+90,120.05599212646484
+91,212.05149841308594
+92,1128.4677734375
+93,179.73696899414062
+94,729.4700927734375
+95,624.5028076171875
+96,87.45598602294922
+97,613.9696655273438
+98,465.77850341796875
+99,1176.76953125
+100,1713.817138671875
+101,109.18904113769531
+102,484.53125
+103,223.00367736816406
+104,184.12435913085938
+105,172.3700408935547
+106,87.6129379272461
+107,364.3734130859375
+108,508.0361328125
+109,77.96926879882812
+110,116.11329650878906
+111,266.626220703125
+112,202.28152465820312
+113,376.0977783203125
+114,127.93818664550781
+115,838.083984375
+116,66.45230102539062
+117,247.91183471679688
+118,424.33404541015625
+119,217.76974487304688
+120,1662.1240234375
+121,748.4871215820312
+122,102.93913269042969
+123,4314.5400390625
+124,204.0469512939453
+125,109.38993835449219
+126,35079.82421875
+127,1404.8218994140625
+128,952.2924194335938
+129,725.766357421875
+130,901.718505859375
+131,327.2513732910156
+132,767.4285888671875
+133,619.0957641601562
+134,927.028564453125
+135,347.1879577636719
+136,232.33673095703125
+137,222.27906799316406
+138,82.94300842285156
+139,144.34622192382812
+140,179.94741821289062
+141,221.91943359375
+142,661.9664306640625
+143,839.4801635742188
+144,964.8612060546875
+145,919.7672729492188
+146,98.85664367675781
+147,3154.924072265625
+148,737.999755859375
+149,113.01490783691406
+150,4807.203125
+151,469.78363037109375
+152,416.414794921875
+153,363.02398681640625
+154,247.61309814453125
+155,297.6471252441406
+156,150.1171417236328
+157,573.760986328125
+158,188.0963134765625
+159,305.11083984375
+160,469.571044921875
+161,1141.966552734375
+162,282.51947021484375
+163,640.5198974609375
+164,251.78041076660156
+165,3304.4404296875
+166,324.857666015625
+167,1349.5191650390625
+168,337.8106689453125
+169,788.873779296875
+170,34615.78125
+171,1620.6519775390625
+172,3106.46435546875
+173,443.064453125
+174,558.744140625
+175,2997.193115234375
+176,96.51470184326172
+177,2576.751953125
+178,300.2566223144531
+179,476.29315185546875
+180,502.770751953125
+181,54234.55078125
+182,492.4431457519531
+183,829.4091796875
+184,3383.392578125
+185,151.7763214111328
+186,2096.01904296875
+187,212.9459991455078
+188,3276.21142578125
+189,1330.978759765625
+190,1938.218505859375
+191,413.1387939453125
+192,480.37872314453125
+193,418.84039306640625
+194,805.3323364257812
+195,464.34356689453125
+196,2148.400390625
+197,602.6488647460938
+198,417.49365234375
+199,2744.437255859375
+200,69.17740631103516
+201,7419.19189453125
+202,259.41253662109375
+203,236.95880126953125
+204,1396.227783203125
+205,560.71240234375
+206,676.3497314453125
+207,1341.9593505859375
+208,383.1351318359375
+209,508.0167236328125
+210,693.378173828125
+211,442.42578125
+212,317.98101806640625
+213,253.62765502929688
+214,603.0238037109375
+215,1498.3974609375
+216,2047.898193359375
+217,1206.361083984375
+218,854.2302856445312
+219,1257.2398681640625
+220,876.58642578125
+221,468.96343994140625
+222,287.5798034667969
+223,1048.658935546875
+224,3740.224609375
+225,380.68096923828125
+226,1930.039794921875
+227,945.2046508789062
+228,870.4012451171875
+229,207.75
+230,1282.3114013671875
+231,170.57391357421875
+232,34035.11328125
+233,824.1802978515625
+234,1321.4935302734375
+235,544.4994506835938
+236,418.048828125
+237,245.936767578125
+238,288.75482177734375
+239,376.4176940917969
+240,396.01513671875
+241,421.2037048339844
+242,1241.68505859375
+243,581.6722412109375
+244,1282.613525390625
+245,1006.0900268554688
+246,600.6984252929688
+247,398.2529296875
+248,35698.90234375
+249,1157.1995849609375
+250,773.305908203125
+251,552.0653686523438
+252,1699.5013427734375
+253,283.8804626464844
+254,562.9462280273438
+255,667.875
+256,466.95709228515625
+257,399.7000732421875
+258,474.777099609375
+259,1725.865966796875
+260,310.0705871582031
+261,199.14935302734375
+262,338.11907958984375
+263,308.2383117675781
+264,405.7358703613281
+265,680.1403198242188
+266,119.51339721679688
+267,197.7222900390625
+268,538.3505859375
+269,612.3535766601562
+270,2835.82861328125
+271,2891.37548828125
+272,461.57391357421875
+273,609.6471557617188
+274,229.94732666015625
+275,3409.61181640625
+276,37065.28515625
+277,999.9266357421875
+278,27983.87890625
+279,301.8092346191406
+280,1595.364990234375
+281,838.091064453125
+282,3710.05224609375
+283,1089.48046875
+284,1759.3096923828125
+285,1162.30322265625
+286,444.10955810546875
+287,3477.90234375
+288,1392.64013671875
+289,829.9650268554688
+290,442.4349060058594
+291,140.54620361328125
+292,1919.3917236328125
+293,510.51446533203125
+294,2170.2197265625
+295,1404.8033447265625
+296,311.72698974609375
+297,440.96160888671875
+298,587.8609619140625
+299,483.2724914550781