scripts/train_release_v0_1.sh

#!/usr/bin/env bash
set -euo pipefail

# Reproduce the final Stage 46 continuation. Run from the repository root after
# installing Isaac Lab and this extension. CUDA_VISIBLE_DEVICES may be changed.
export CUDA_VISIBLE_DEVICES="${CUDA_VISIBLE_DEVICES:-0}"
export DROPBEAR_FEET_MIN_DISTANCE=0.16
export DROPBEAR_FEET_MIN_LATERAL_SEPARATION=0.10

warm_start_run="logs/rsl_rl/dropbear_velocity/release_v0.1.0_warm_start"
mkdir -p "${warm_start_run}"
install -m 0644 \
  checkpoints/dropbear_locomotion_v0.1.0/warm_start_model_8666.pt \
  "${warm_start_run}/model_8666.pt"

python -u scripts/train_dropbear.py \
  --task Isaac-Dropbear-Velocity-v0 \
  --headless \
  --device cuda:0 \
  --num_envs 512 \
  --seed 42 \
  --resume \
  --load_run release_v0.1.0_warm_start \
  --checkpoint model_8666.pt \
  --max_iterations 80 \
  --save-interval 5 \
  --run_name release_v0.1.0_stage46 \
  --actor-update-scale 0.04 \
  --learning-rate 3e-5 \
  --ppo-schedule fixed \
  --entropy-coef 0.0001 \
  --tracking-std 0.10 \
  --yaw-tracking-weight 0.5 \
  --yaw-tracking-std 0.5 \
  --fixed-forward-speed 0 \
  --fixed-lateral-speed 0.20 \
  --fixed-yaw-rate 0 \
  --standing-env-fraction 0 \
  --stable-forward-weight 2 \
  --stable-forward-std 0.10 \
  --com-control \
  --com-stand-height 1.052 \
  --com-height-delta 0 \
  --com-height-error-scale 0.03 \
  --com-vertical-velocity-error-scale 0.08 \
  --com-velocity-weight 12 \
  --com-planar-velocity-penalty-weight -40 \
  --com-planar-velocity-forward-scale 1 \
  --com-planar-velocity-lateral-scale 2 \
  --com-height-weight 1 \
  --com-position-weight 0 \
  --com-velocity-xy-std 0.07 \
  --com-velocity-z-std 0.05 \
  --com-height-std 0.02 \
  --com-position-std 0.2 \
  --gait-period 0.55 \
  --arm-swing-weight 1 \
  --arm-swing-amplitude 0.2 \
  --arm-swing-std 0.18 \
  --arm-extension-penalty-weight -2 \
  --arm-extension-soft-limit 0.35 \
  --arm-counterweight-penalty-weight -5 \
  --arm-counterweight-soft-limit 0.1 \
  --foot-phase-velocity-weight 1.5 \
  --swing-foot-speed-factor 2 \
  --swing-foot-forward-std 0.2 \
  --swing-foot-lateral-std 0.15 \
  --feet-self-interaction-penalty-weight -120 \
  --feet-min-distance 0.16 \
  --feet-min-lateral-separation 0.10 \
  --feet-approach-penalty-weight -18 \
  --feet-approach-distance 0.20 \
  --feet-approach-lateral-separation 0.14 \
  --feet-closing-speed-scale 0.20 \
  --feet-touchdown-clearance-weight 4 \
  --feet-late-swing-phase-start 0.80 \
  --feet-touchdown-distance 0.16 \
  --feet-touchdown-lateral-separation 0.10 \
  --feet-touchdown-distance-std 0.05 \
  --feet-touchdown-lateral-std 0.04 \
  --actor-base-lin-vel \
  --plane-only \
  --disable-pushes \
  --reset-policy-joints-only \
  --reset-joint-position-range 1 1 \
  --reset-joint-velocity-range 0 0