### Role
You are an expert high-level planner for a 2D-gird game. You must strictly adhere to the requirements outlined in the system instructions and tasks I have provided to you.

### Instructions
1. Your task is to plan a feasible sequence of moves in a {{world_height}}×{{world_length}} grid environment, from a start state to a goal state.  
2. The plan should be defined by a series of key anchor coordinates.  
3. You must identify exactly {{num_anchors}} feasible anchor states for the given task. These should be important turning points or subgoals used to navigate around obstacles or toward objectives.  

### Anchor State Selection Strategy
- The plan does not need to be the shortest; the priority is feasibility and safety (avoiding all obstacles).  
- Anchor states should be strategically located at key subgoals: clearing an obstacle, moving around blocking tiles, or forming partial progress toward objectives.  
- Explore multiple valid strategies and select one reasonable plan.  

### Reward Context
- Rewards are given as follows:  
  {{reward_design}}  
  {{reward_feedback}}  
- You are also given information about your previous attempt:  
  - Actions generated: {{total_actions[list(objective_tile_dict.keys())[i]]}}  
  - Start position: {{prev_protagonist_position}}  
  - End position: {{protagonist_position}}  
  - Distance from objective: {{distance_from_objective}}  
  - Objective location: {{list(objective_tile_dict.values())[i]}}  
  - GTB Reward received: {{reward_this_objective[list(objective_tile_dict.keys())[i]]}}  

### Output Format
- Strictly output the result in the following format, without any explanation:  
<trajectory for planning> = [(start_x, start_y), (anchor_1_x, anchor_1_y), ..., (end_x, end_y)]
---
### Examples

**Example 1:**
Task: You are in a 10 by 10 world. There are obstacles that you have to avoid at: (4,7), (8,6), (3,3), (9,5), (8,9), (1,1), (5,4), (1,3), (9,9), (4,1), (5,7), (1,6), (9,0), (8,3), (0,0), (7,1), (4,6), (5,0), (2,5) and (4,0). Go from (2,1) to (0,2).
<trajectory for planning> = [(2,1),(2,2),(0,2)]

**Example 2:**
Task: You are in a 10 by 10 world. There are obstacles that you have to avoid at: (0,7), (3,2), (0,4), (3,4), (4,6), (7,2), (7,3), (2,0), (3,9), (9,3), (8,2), (9,5), (8,4), (7,5), (4,8), (5,2), (5,5), (7,8), (6,3) and (9,8). Go from (6,8) to (6,1). 
<trajectory for planning> = [(6,0),(6,4),(5,3),(4,2),(6,1)]

**Example 3:**
Task: You are in a 10 by 10 world. There are obstacles that you have to avoid at: (8,5), (7,2), (1,7), (2,0), (3,2), (5,0), (1,9), (3,3), (3,6), (4,7), (0,3), (5,7), (5,3), (4,6), (2,8), (4,3), (9,0), (7,5), (5,5) and (8,9). Go from (0,8) to (7,1).
<trajectory for planning> = [(0,8),(0,4),(1,1),(7,1)]
---
### Task to Solve
Task: {{task}}
<trajectory for planning> =