JOB ID 28258
Autonomous Driving Fusion Algorithm Engineer (MOT / Multi-Sensor Fusion)
- メーカー, 自動車
- 製品エンジニア（ハード・ソフトウェア）
- 東京都
- 1000万円～2000万円
2024年4月に設立された日本のディープテック系スタートアップ企業。大型トラック向け自動運転システムの開発を専門とし、ドライバー不足や高齢化、物流需要の増加といった日本の物流業界が抱える課題の解決を目指している。高速道路を利用した幹線輸送に重点を置き、AIを活用した先進的な自動運転技術により、輸送の安全性と業務効率の向上、物流コストの削減を推進している。自動車メーカーや物流企業との連携を通じて、自動運転トラックの実用化と次世代の物流インフラの構築に取り組んでいる。
A Japanese deep-tech startup established in April 2024, specializing in the development of autonomous driving systems for heavy-duty trucks. The company aims to transform Japan’s logistics industry by addressing challenges such as driver shortages, an aging workforce, and increasing freight transportation demand. Focusing on autonomous highway transportation, it develops advanced AI-powered technologies designed to improve transportation safety, operational efficiency, and logistics sustainability. Through collaboration with automotive manufacturers and logistics companies, the company is working toward the commercialization of autonomous trucking solutions and the development of next-generation automated freight transportation infrastructure in Japan.
This role focuses on multi-object tracking (MOT) and multi-sensor fusion for autonomous driving perception, covering both object-level fusion and early fusion. The position is responsible for advancing data association and state estimation algorithms, cross-sensor fusion architecture, evaluation methodology, and the deployment of robust, production-grade tracking and fusion modules on large-scale real-world driving data
募集要項
-
- 職種要約
- - Design and iterate on multi-object tracking (MOT) algorithms for autonomous driving, including data association, state estimation (KF/EKF/UKF/IMM, etc.), track lifecycle management, and handling of occlusions and ID switches, to improve tracking stability in complex urban and highway scenarios.
- Develop object-level fusion algorithms that perform spatio-temporal alignment, cross-sensor association, and confidence fusion of detection results from Camera / LiDAR / Radar, producing a unified and stable object list.
- Participate in the design and deployment of early fusion solutions, including spatio-temporal synchronization of multi-modal data and alignment/fusion strategies for raw or feature-level data.
- Analyze fusion and tracking bad cases from real-world road tests and simulation data, build evaluation metrics and automated benchmarking pipelines, and continuously improve algorithm performance.
- Own the engineering, performance optimization, and mass-production delivery of algorithms on embedded automotive platforms.
- Collaborate closely with perception, localization, calibration, planning & control, and data platform teams to define interface specifications for the fusion module.
-
- 応募資格（必須）
- Requirements
- Master's degree or above in Computer Science, Automation, Electronic Engineering, Control, Robotics, or a related field.
- 2+ years of R&D experience in perception/fusion algorithms for autonomous driving or robotics, with a solid foundation in probability theory, state estimation, and optimization.
- Deep understanding of multi-object tracking and multi-sensor fusion methods; familiar with the Kalman filter family, IMM, and data association algorithms such as JPDA, MHT, and the Hungarian algorithm.
- Familiar with the characteristics and error models of Camera, LiDAR, and Radar sensors; experienced in spatio-temporal synchronization and extrinsic calibration.
- Proficient in C++ and familiar with Python, with strong coding standards and engineering skills; experienced with common libraries such as Eigen, PCL, and OpenCV.
- Familiar with ROS/ROS2 or autonomous driving middleware (e.g., Apollo, Autoware, CyberRT) development workflows.
- Strong analytical and problem-solving skills, good communication and teamwork, and a strong sense of ownership.
-
- 応募資格（優遇）
- Preferred Qualifications
- Hands-on MOT project experience (e.g., SORT, AB3DMOT), or ownership of the tracking module in a mass-production program.
- Mass-production delivery experience with Camera + LiDAR + Radar object-level fusion.
- Research or deployment experience in early fusion.
- SLAM project experience (e.g., LIO-SAM, FAST-LIO, VINS) and familiarity with tightly coupled state estimation.
- Publications at robotics/autonomous-driving conferences such as ICRA or IROS, or strong results on public tracking/detection benchmarks such as nuScenes, Waymo, or KITTI.
-
- 勤務地
- 東京都
-
- 勤務スタイル
- 在宅勤務OK, フレックス勤務OK
-
- 給与
- 1000万円～2000万円
-
- こだわり条件
- 海外展開, 土日祝休み, 大手企業
Consultation
自動車・モビリティ業界における
転職やキャリア設計、
求人の選定の
ご相談を承ります。
