EmbodiedAI
/
RoboWaiter
mirror of https://github.com/HPCL-EI/RoboWaiter.git
3
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

查询函数融入场景中

This commit is contained in:
ChenXL97 2023-11-18 22:30:14 +08:00
parent 1033b95781
commit 8bdf5a96b4
6 changed files with 81 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
robowaiter/behavior_lib/act/DealChat.py
View File

@ -9,7 +9,8 @@ class DealChat(Act):
self.chat_history = ""
self.function_success = False
self.func_map = {
"create_sub_task": self.create_sub_task
"create_sub_task": self.create_sub_task,
"get_object_info": self.get_object_info
}
def _update(self) -> ptree.common.Status:
@ -53,3 +54,19 @@ class DealChat(Act):
print("参数解析错误")
self.scene.robot.expand_sub_task_tree(goal_set)
def get_object_info(self,**args):
try:
obj = args['obj']
self.function_success = True
except:
obj = None
print("参数解析错误")
near_object = "None"
if obj == "洗手间":
near_object = "大门"
return near_object

7
robowaiter/llm_client/data/fix_questions.txt
View File

@ -36,3 +36,10 @@ create_sub_task
抱歉，我这就去开空调和关窗帘。
create_sub_task
{"goal":"Is(AC,On),Is(Curtain,Off)"}
请问洗手间在哪里？
洗手间在这附近
get_object_info
{"obj":"洗手间"}

37
robowaiter/llm_client/multi_rounds.py
View File

@ -29,16 +29,30 @@ root_path = get_root_path()
# load test questions
file_path = os.path.join(root_path,"robowaiter/llm_client/data/fix_questions.txt")
functions = get_tools()
fix_questions_dict = {}
no_reply_functions = ["create_sub_task"]
with open(file_path,'r',encoding="utf-8") as f:
#读取所有行
lines = f.read().strip()
sections = re.split(r'\n\s*\n', lines)
for s in sections:
x = s.split()
fix_questions_dict[x[0]] = x[1:]
if len(x) == 2:
fix_questions_dict[x[0]] = {
"answer": x[1],
"function": None
}
else:
fix_questions_dict[x[0]] = {
"answer": x[1],
"function": x[2],
"args": x[3]
}
functions = get_tools()
role_system = [{
"role": "system",
"content": "你是RoboWaiter,一个由HPCL团队开发的机器人服务员，你在咖啡厅工作。接受顾客的指令并调用工具函数来完成各种服务任务。如果顾客问你们这里有什么，或者想要点单，你说我们咖啡厅提供咖啡，水，点心，酸奶等食物。如果顾客不需要你了，你就回到吧台招待。如果顾客叫你去做某事，你回复：好的，我马上去做这件事。",
@ -55,11 +69,12 @@ def new_response():
def parse_fix_question(question):
response = new_response()
fix_ans = fix_questions_dict[question]
if len(fix_ans)<=1: #简单对话
message = {'role': 'assistant', 'content': fix_ans[0], 'name': None,
if not fix_ans['function']: #简单对话
message = {'role': 'assistant', 'content': fix_ans["answer"], 'name': None,
'function_call': None}
else:
reply, func,args = fix_ans
func = fix_ans["function"]
args = fix_ans["args"]
# tool_response = dispatch_tool(function_call["name"], json.loads(args))
# logger.info(f"Tool Call Response: {tool_response}")
message = {'role': 'assistant',
@ -115,7 +130,7 @@ def deal_response(response, history, func_map=None ):
}
history.append(t)
return function_call["name"], return_message
return function_call["name"], tool_response
else:
return_message = response["choices"][0]["message"]
@ -133,8 +148,14 @@ def ask_llm(question,history, func_map=None, retry=3):
function_call,result = deal_response(response, history, func_map)
if function_call:
if question in fix_questions_dict:
reply = fix_questions_dict[question][0]
result = single_round(reply,"你是机器人服务员，请把以下句子换一种表述方式对顾客说，但是意思不变，尽量简短：\n")
if fix_questions_dict[question]['function'] in no_reply_functions:
reply = fix_questions_dict[question]["answer"]
result = single_round(reply,
"你是机器人服务员，请把以下句子换一种表述方式对顾客说，但是意思不变，尽量简短：\n")
else:
reply = fix_questions_dict[question]["answer"]
result = single_round(f"你是机器人服务员，顾客想知道{question}, 你的具身场景查询返回的是{result},请把按照以下句子对顾客说，{reply}, 尽量简短。\n")
message = {'role': 'assistant', 'content': result, 'name': None,
'function_call': None}
history.append(message)

50
robowaiter/llm_client/tool_register.py
View File

@ -158,31 +158,31 @@ def stop_serve(
# @register_tool
# def get_object_info(
# obj: Annotated[str, '需要获取信息的物体名称', True]
# ) -> str:
# """
# 获取场景中指定物体 `object` 在哪里，不涉及到具体的执行任务
# 如果`object` 是一个地点，例如洗手间，则输出大门。
# 如果`object`是咖啡，则输出桌子，咖啡在桌子上。
# 如果`object` 是空桌子，则输出一号桌
# """
# near_object = None
# # if obj == "Table":
# # near_object = "Bar"
# # if obj == "洗手间":
# # near_object = "大门"
# # if obj == "空桌子":
# # near_object = "一号桌"
# if obj in find_obj_utils.all_loc: # object是一个地点
# mp = list(find_obj_utils.loc_map[obj])
# # near_object = random.choice(mp)
# near_object = mp
# if obj in find_obj_utils.all_obj: # object是一个物品
# near_ls = find_obj_utils.all_loc + find_obj_utils.all_obj
# near_object = random.choices(near_ls,k=5)
# return near_object
@register_tool
def get_object_info(
obj: Annotated[str, '需要获取信息的物体名称', True]
) -> str:
"""
获取场景中指定物体 `object` 在哪里，不涉及到具体的执行任务
如果`object` 是一个地点，例如洗手间，则输出大门。
如果`object`是咖啡，则输出桌子，咖啡在桌子上。
如果`object` 是空桌子，则输出一号桌
"""
near_object = None
# if obj == "Table":
# near_object = "Bar"
if obj == "洗手间":
near_object = "大门"
# if obj == "空桌子":
# near_object = "一号桌"
if obj in find_obj_utils.all_loc: # object是一个地点
mp = list(find_obj_utils.loc_map[obj])
# near_object = random.choice(mp)
near_object = mp
if obj in find_obj_utils.all_obj: # object是一个物品
near_ls = find_obj_utils.all_loc + find_obj_utils.all_obj
near_object = random.choices(near_ls,k=5)
return near_object
# @register_tool
# def find_location(

5
robowaiter/scene/tasks/GQA.py → robowaiter/scene/tasks/GQA/GQA.py
View File

@ -15,9 +15,8 @@ class SceneGQA(Scene):
def __init__(self, robot):
super().__init__(robot)
# 在这里加入场景中发生的事件， (事件发生的时间，事件函数)
self.event_list = [
(5, self.create_chat_event("哪里有空桌子")),
(12, self.create_chat_event("可以带我去吗")),
self.new_event_list = [
(3, self.customer_say, ("System","请问洗手间在哪里？"))
]
def _reset(self):

0
robowaiter/scene/tasks/GQA/__init__.py Normal file
View File