Information collecting device, information providing device, information collecting method, and information providing method
Patent Information
- Application Number
- JP2023201124
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-11-28
- Publication Date
- 2025-06-09
AI Technical Summary
【0030】 本発明によれば、時系列データを抽出する精度の向上を図ることができる。
Smart Images

Figure 2025086821000001_ABST
Abstract
Claims
1. An information collection device comprising a control unit, wherein the control unit is configured to: acquire time-series data indicating state transitions of one or more agents; convert the acquired time-series data into a feature vector using an encoder configured to encode by incorporating information from multiple time points; and associate the obtained feature vector with the time-series data and store it in a database, and is configured to execute the above operations, an information collection device.
2. The time-series data is configured to indicate state transitions of multiple agents, and the encoder is configured to further incorporate spatial information for encoding, The information collection device according to Claim 1.
3. The encoder is configured to be trained to predict the future states of one or more agents from their state transitions, thereby encoding by incorporating information from multiple time points, The information collection device according to Claim 1.
4. The time-series data includes trajectory data, The information collection device according to Claim 1.
5. The time-series data includes moving image data, The information collection device according to Claim 1.
6. The control unit is further configured to acquire attribute information of the one or more agents, and converting the acquired time-series data into a feature vector is achieved by converting the acquired time-series data and the attribute information of the one or more agents into a feature vector, The information collection device according to Claim 1.
7. The time-series data is further configured to indicate state transitions of one or more objects, and the encoder is configured to further incorporate spatial information for encoding, The information collection device according to Claim 1.
8. The control unit is further configured to acquire attribute information of the one or more objects, and converting the acquired time-series data into a feature vector is achieved by converting the acquired time-series data and the attribute information of the one or more objects into a feature vector, The information collection device according to Claim 7.
9. The control unit is further configured to acquire environmental information regarding the environment in which the one or more agents exist, and converting the acquired time-series data into a feature vector is achieved by converting the acquired time-series data and the environmental information into a feature vector, The information collection device according to Claim 1.
10. The agent is a robot device or a living organism, The information collection device according to claim 1.
11. An information providing device connected to a database storing feature vectors associated with time series data and including a control unit, The time series data indicates state transitions of one or more agents, The feature vector is obtained by converting the time series data using an encoder configured to encode by incorporating information at a plurality of times, The control unit, Receiving a query for the database from a requester, Extracting, as related feature vectors, feature vectors that match the query from the database, and Returning, as related time series data, the time series data associated with the extracted related feature vectors to the requester, Is configured to execute, Information providing device.
12. Receiving the query is constituted by receiving a plurality of the queries, Extracting the related feature vectors is constituted by extracting a plurality of related feature vectors respectively matching each of the plurality of queries, Regarding the plurality of related feature vectors to be extracted, it is allowed that the information sources of the time series data associated with some of the related feature vectors are different from the information sources of the time series data associated with the other related feature vectors, The information providing device according to claim 11.
13. Regarding the feature vector to be extracted, it is allowed that the types of the one or more agents in the related time series data are different from the types of the one or more target agents in the task targeted by the query, The information providing device according to claim 11.
14. The query is generated by encoding target time series data with the encoder, The scale of the target time series data has been changed at the requester, The information providing device according to claim 11.
15. Receiving the query from the requester, Receiving target time series data from the requester, Changing the scale of the target time series data, and Generating the query by encoding the changed target time series data with the encoder, Including, The information providing device according to claim 11.
16. Returning the time-series data as related time-series data to the requester includes changing the scale of the related time-series data, and returning the changed related time-series data to the requester, and includes the information providing apparatus according to claim 11.
17. A computer acquires time-series data indicating state transitions of one or more agents, uses an encoder configured to encode by incorporating information at a plurality of times to convert the acquired time-series data into a feature vector, and stores the obtained feature vector in association with the time-series data in a database, and executes an information collection method.
18. An information providing method by a computer connected to a database storing feature vectors associated with time-series data, wherein the time-series data indicates state transitions of one or more agents, the feature vector is obtained by converting the time-series data using an encoder configured to encode by incorporating information at a plurality of times, the computer receives a query for the database from a requester, extracts a feature vector matching the query as a related feature vector from the database, and returns time-series data associated with the extracted related feature vector as related time-series data to the requester, and executes an information providing method.