{"id":{"repo_id":"bu","oai_identifier":"oai:open.bu.edu:2144/49900"},"canonical_url":"https://search.ndltd.org/etd/bu/oai:open.bu.edu:2144/49900","repository":{"repo_id":"bu","name":"Boston University","base_url":"https://open.bu.edu/oai/request"},"display":{"title":"Learning spatial representations for efficient robot navigation","abstract":"How can a robot learn to understand and navigate its 3D environment? This research proposes a method for robots to comprehend and traverse their 3D surroundings along a specific route without relying on large architectures or costly 3D sensors like LiDAR, which are impractical for autonomous robots due to computational limitations and excessive power consumption. Instead, it suggests rectifying navigation commands predicted by smaller Imitation Learning models using Reinforcement Learning. The research examines various navigation reward functions used by other studies and introduces a novel multiplicative reward that outperforms them in guiding a robot from a designated starting point through a crowded path to its destination. It also explores motion prediction algorithms based on past motion sequences to anticipate the movements of pedestrians around the robot for facilitating the reinforcement learning algorithm. Moreover, it introduces a custom-designed CARLA RL environment for robot navigation, beneficial for future research in the field. Through experimentation, the research demonstrates that RL effectively corrects and predicts precise actions for the robot to reach its destination, eliminating the necessity for deploying large models for optimal performance.","abstract_html":"How can a robot learn to understand and navigate its 3D environment? This research proposes a method for robots to comprehend and traverse their 3D surroundings along a specific route without relying on large architectures or costly 3D sensors like LiDAR, which are impractical for autonomous robots due to computational limitations and excessive power consumption. Instead, it suggests rectifying navigation commands predicted by smaller Imitation Learning models using Reinforcement Learning. The research examines various navigation reward functions used by other studies and introduces a novel multiplicative reward that outperforms them in guiding a robot from a designated starting point through a crowded path to its destination. It also explores motion prediction algorithms based on past motion sequences to anticipate the movements of pedestrians around the robot for facilitating the reinforcement learning algorithm. Moreover, it introduces a custom-designed CARLA RL environment for robot navigation, beneficial for future research in the field. Through experimentation, the research demonstrates that RL effectively corrects and predicts precise actions for the robot to reach its destination, eliminating the necessity for deploying large models for optimal performance.","abstract_has_math":false,"creators":["Sengupta, Kathakoli"],"institution":"Boston University","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":[],"advisors":["Ohn Bar, Eshed"],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024","date_published":"2024","updated_at":"2026-11-18T19:40:34Z","subjects":["Artificial intelligence","Generative models","Imitation learning","Motion prediction","Reinforcement learning","Reward design","Robot navigation"],"languages":["en_US"],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2144/49900","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/2144/49900","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.bu.edu/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aopen.bu.edu%3A2144%2F49900","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ohn Bar, Eshed"]},{"key":"dc:creator","label":"Author","values":["Sengupta, Kathakoli"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-03-06T16:06:48Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-03-06T16:06:48Z"]},{"key":"dc:date.issued","label":"Date","values":["2024"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2025-03-06T02:01:05Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis/Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Boston University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial intelligence","Generative models","Imitation learning","Motion prediction","Reinforcement learning","Reward design","Robot navigation"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2144/49900"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["2024"]},{"key":"dc:description.abstract","label":"Abstract","values":["How can a robot learn to understand and navigate its 3D environment? This research proposes a method for robots to comprehend and traverse their 3D surroundings along a specific route without relying on large architectures or costly 3D sensors like LiDAR, which are impractical for autonomous robots due to computational limitations and excessive power consumption. Instead, it suggests rectifying navigation commands predicted by smaller Imitation Learning models using Reinforcement Learning. The research examines various navigation reward functions used by other studies and introduces a novel multiplicative reward that outperforms them in guiding a robot from a designated starting point through a crowded path to its destination. It also explores motion prediction algorithms based on past motion sequences to anticipate the movements of pedestrians around the robot for facilitating the reinforcement learning algorithm. Moreover, it introduces a custom-designed CARLA RL environment for robot navigation, beneficial for future research in the field. Through experimentation, the research demonstrates that RL effectively corrects and predicts precise actions for the robot to reach its destination, eliminating the necessity for deploying large models for optimal performance."]},{"key":"dc:title","label":"Title","values":["Learning spatial representations for efficient robot navigation"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ohn Bar, Eshed"],"dc:creator":["Sengupta, Kathakoli"],"dc:date.accessioned":["2025-03-06T16:06:48Z"],"dc:date.available":["2025-03-06T16:06:48Z"],"dc:date.issued":["2024"],"dc:date.updated":["2025-03-06T02:01:05Z"],"dc:description":["2024"],"dc:description.abstract":["How can a robot learn to understand and navigate its 3D environment? This research proposes a method for robots to comprehend and traverse their 3D surroundings along a specific route without relying on large architectures or costly 3D sensors like LiDAR, which are impractical for autonomous robots due to computational limitations and excessive power consumption. Instead, it suggests rectifying navigation commands predicted by smaller Imitation Learning models using Reinforcement Learning. The research examines various navigation reward functions used by other studies and introduces a novel multiplicative reward that outperforms them in guiding a robot from a designated starting point through a crowded path to its destination. It also explores motion prediction algorithms based on past motion sequences to anticipate the movements of pedestrians around the robot for facilitating the reinforcement learning algorithm. Moreover, it introduces a custom-designed CARLA RL environment for robot navigation, beneficial for future research in the field. Through experimentation, the research demonstrates that RL effectively corrects and predicts precise actions for the robot to reach its destination, eliminating the necessity for deploying large models for optimal performance."],"dc:identifier.uri":["https://hdl.handle.net/2144/49900"],"dc:language.iso":["en_US"],"dc:subject":["Artificial intelligence","Generative models","Imitation learning","Motion prediction","Reinforcement learning","Reward design","Robot navigation"],"dc:title":["Learning spatial representations for efficient robot navigation"],"dc:type":["Thesis/Dissertation"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Boston University"]},"updated_at":"2026-11-18T19:40:34Z"}