Fix robot initial positions: use polar coordinate generation on outer circle

[Copilot]

Robot dogs spawned at fixed position [0.0, 0.6, 0.5] (0.6m from center), violating the requirement for random distribution on outer circle at 3.0m radius. This prevented proper testing of the outer→inner→center navigation path.

Changes

Configuration (cfg.py)

Added arena geometry parameters to VBotSection001EnvCfg:

arena_outer_radius: float = 3.0  # Outer circle radius
arena_inner_radius: float = 1.5  # Inner circle radius  
boundary_radius: float = 3.5     # Physical boundary
arena_center: list = field(default_factory=lambda: [0.0, 0.0])

Position Generation (vbot_section001_np.py)

Replaced fixed position with polar coordinate random generation in reset():

Before:

robot_init_pos = np.tile(cfg.init_state.pos, (num_envs, 1))

After:

# Polar coordinate random generation on outer circle
robot_init_xy = np.zeros((num_envs, 2), dtype=np.float32)
for i in range(num_envs):
    theta = np.random.uniform(0, 2 * np.pi)
    radius = cfg.arena_outer_radius + np.random.uniform(-0.1, 0.1)
    robot_init_xy[i, 0] = radius * np.cos(theta)
    robot_init_xy[i, 1] = radius * np.sin(theta)

robot_init_xy += np.array(cfg.arena_center, dtype=np.float32)
robot_init_pos = np.column_stack([robot_init_xy, np.full(num_envs, 0.5, dtype=np.float32)])

Testing

Added test_initial_position_generation.py with 8 unit tests validating:

• Distance from center: 2.9-3.1m
• Position uniqueness across resets
• Angular distribution coverage
• Fixed height at 0.5m

Result

Dogs now spawn randomly distributed on outer circle (2.9-3.1m from center) with different angles on each reset, enabling proper navigation path testing per ACCEPTANCE_CHECKLIST.md requirements.

Original prompt

问题描述

当前 VBot 导航竞技场任务中，机器狗的初始位置生成不符合题目要求：

现状 ❌

• 所有机器狗都从固定位置 pos = [0.0, 0.6, 0.5] 出生
• 这违反了"初始位置随机生成"的要求
• 机器狗几乎就在圆心附近，无法真正测试外圈→内圈→圆心的导航路径

要求 ✅

根据 ACCEPTANCE_CHECKLIST.md 和题目规范：

• 10只机器狗应在外圈（半径 3.0m）上随机分布
• 每次重置产生不同的随机位置
• 使用极坐标随机生成：
  • 角度 θ: [0, 2π] 均匀随机
  • 半径 r: 3.0 ± 0.1m（外圈范围）
  • 高度 z: 0.5m（固定）

影响范围

• 文件: motrix_envs/src/motrix_envs/navigation/vbot/vbot_section001_np.py
• 方法: reset() 中的初始位置生成逻辑（当前 L766-787）
• 相关配置: motrix_envs/src/motrix_envs/navigation/vbot/cfg.py

修复内容

修改 reset() 方法中的位置初始化代码，将其从固定位置改为在外圈随机分布：

# 原代码（固定位置）:
robot_init_pos = np.tile(cfg.init_state.pos, (num_envs, 1))

# 新代码（极坐标随机生成在外圈）:
robot_init_xy = np.zeros((num_envs, 2), dtype=np.float32)
for i in range(num_envs):
    theta = np.random.uniform(0, 2 * np.pi)
    radius = cfg.arena_outer_radius + np.random.uniform(-0.1, 0.1)
    robot_init_xy[i, 0] = radius * np.cos(theta)
    robot_init_xy[i, 1] = radius * np.sin(theta)

# 添加圆心偏移（如果存在）
robot_init_xy += cfg.arena_center

# 构造完整的XYZ位置
robot_init_pos = np.column_stack([robot_init_xy, np.full(num_envs, 0.5, dtype=np.float32)])

验证标准

• 初始位置在外圈（距圆心 ~3.0m）
• 每次重置位置随机不同
• 10只狗位置分散（角度不同）
• 符合 test_arena_navigation.py 预期输出

This pull request was created from Copilot chat.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.